flooding to financial disaster: an introduction to extreme...

44
Flooding to Financial Disaster: An Introduction to Extreme Value Theory Grant Weller Department of Statistics Colorado State University Joint work with: Dan Cooley, CSU Steve Sain, NCAR

Upload: others

Post on 19-Oct-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Flooding to Financial Disaster:An Introduction to Extreme Value Theory

Grant WellerDepartment of StatisticsColorado State University

Joint work with:Dan Cooley, CSUSteve Sain, NCAR

1

Page 2: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Extreme Events

Q: What does it mean to be an ‘extreme event’ ?

2

Page 3: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Extreme Events

Q: What does it mean to be an ‘extreme event’ ?

A: It depends whom you ask:

• Financial analysts: ‘shocks’ to a time series (market crash)

• Insurance companies: costly (though perhaps not rare)phenomena (hurricanes)

• Meteorologists: rare weather phenomena (a brown Christ-mas in Moorhead?)

• Applied mathematicians: extremes of Gaussian/other pro-cesses, large deviations

• Statisticians (like me!): Extreme Value Theory (this talk:data perspective)

• Many possible answers!

3

Page 4: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Extreme Value Theory

“Ordinary” Statistics: try to describe a distribution of data;possibly ignore very large or small values (outliers)

Extreme Value Theory: try to characterize the tail of thedistribution; uses only the extreme observations

“Ordinary” Stats

����

Extremes

@@R

4

Page 5: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why study extremes?

While infrequent, extremes often have large human impact.

Goal of an extreme value analysis: to quantify the magnitudeof a worst-case really-bad-case scenario.

Application areas:

• hydrology (stream/river flows)

• climate variables: precipitation, wind, heat waves, ...

• finance

• insurance/reinsurance

• engineering (structural design, failure)

• not much done (yet) in medicine, biology, ecology

5

Page 6: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why study extremes?

In fact, you don’t have to go very far to find an example...

Larry Hansen; Fargo Flood homepage

6

Page 7: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why study extremes?

2009 Red River of the North flood

• Highest ever recorded water level at Fargo (40.8 ft on3/28/09) - flood stage is 18 ft

• Tens of millions of dollars in damages (1997: $3.5 billion)

• No classes at Concordia for two weeks!

• Related question: how unlikely was this event?

• We’ll come back to this later...

7

Page 8: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

...first, let’s go back in time

Concordia College 2004-2008Cobber football

• 2007 - school records for totalpoints, touchdowns, yards in aseason

• 2005 & 2007 teams - #1 and#2 in season rush yds and tds

2007 vs. Carleton

Mathematics & Economics major

• Advised by Dr. Zeng

• Other influences: Dr. DougAnderson, Dr. Jim Forde,Dr. Haimeng Zhang, Dr. DanBiebighauser

I had the best undergraduate advisor!

8

Page 9: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Then on to graduate school...

Department of Statistics, CSU Fort Collins, 2008-present

• M.S. in Statistics, 2010

• Ph.D. expected 2013

• Advisor: Dr. Dan Cooley

• learned to ski!

9

Page 10: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Research opportunities

NCAR, Boulder, CO

• Graduate Student Visitor(Summer 2011)

• Climate change research

• Extreme precipitation from cli-mate models - more later

Mesa Lab

SAMSI, RTP, NC

• Visitor for Fall 2011

• Participant in 2011-2012 Un-certainty Quantification pro-gram

• Lived in Chapel Hill, NC

2012 National Champions?

10

Page 11: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Outline

This talk will mainly focus on applications and examples, withvery basic introduction to theoretical results.

• Univariate extremes - Red River flooding example

• Brief introduction to bivariate extremes - describing de-pendence

• Pineapple Express project

• Suggestions for current Concordia math majors

11

Page 12: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

So what about the 2009 Red River flood?

The Fargo Red River station has daily peak flow measure-ments (cfs) going back to 1901.

Q: How ‘rare’ was the 2009 event, in terms of daily flow?

Let’s use springtime (MAM) data from 1901-2008 to try toanswer this.

Two approaches:

1. Analyze all the data; fit a gamma distribution

2. Use only the annual max data; fit a GEV

Warning: Both analyses are likely incorrect, for different rea-sons. Bonus points if you can tell me why later.

12

Page 13: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Modeling all springtime daily data

1900 1920 1940 1960 1980 2000

010000

25000

Red River at Fargo Daily Peak Flow

Year

Dai

ly F

low

(cfs

)

Let Xt be the daily peak flow on day t. Assume that Xt ∼Gamma(α, β) (ignore zero-flow days).

Estimate the parameters via maximum likelihood: α = 0.62,β = 2765.08.

13

Page 14: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Modeling all springtime daily data

Histogram of Daily Flow

Daily Flow

Density

0 5000 10000 15000 20000 25000

0e+00

1e-04

2e-04

3e-04

4e-04

Fit looks ok.

Maximum flow in 2009 was 29100 cfs on March 28th. Whatis the probability that the maximum daily flow in 2009 wouldbe at least this much, assuming this is the right model?

14

Page 15: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Estimate using all daily data

P(Xt > 29100) = 1− FX(29100) = 7.400995× 10−6

P(ann. max > 29100) = 1− P(entire year’s obs < 29100)

= 1− (1− P(one obs > 29100))92

= 1− (1− 7.400995× 10−6)92

= 6.807× 10−4

Associated “return period”: (6.807× 10−4)−1 = 1469 years.

But is this the right way to analyze the data?

15

Page 16: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

All daily data model

Two main problems with this model:

1. Assumes daily river flow rates are iid

2. Underestimates the tail of the distribution

0 5000 10000 15000 20000 25000

05000

10000

15000

20000

25000

Gamma Q-Q Plot

Model

Empirical

99.4% of data and 99.8% of model’s mass are < 15000.

16

Page 17: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Modeling Annual Maxima

Alternative approach: retain only the largest value from eachyear, and fit a generalized extreme value distribution

1900 1920 1940 1960 1980 2000

010000

20000

Red River at Fargo Daily Peak Flow

Year

Dai

ly F

low

(cfs

)

Why?

1. Intuitively, river usually only floods once each year

2. Mathematically, justified by theory

17

Page 18: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Fitting a GEV

Let Mn = max(Xt), t = 1, ..., n. Assume Mn ∼ GEV(µ, σ, ξ).

FMn(x) = P(Mn ≤ x) = exp

{−[1 + ξ

(x− µσ

)]−1/ξ}

PWM estimates: µ = 2367.54, σ = 2496.05, ξ = 0.32.

P(ann. max > 29100) = 1− FMn(29100) = 0.0095

Associated return period: (0.0095)−1 = 105 years

100-year return level: 28561

95% confidence interval (delta method): (8909,48213)

500-year return level: 51536

95% confidence interval (delta method): (1718,101354)

18

Page 19: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

A word of caution

Because we use one observation per year, uncertainty asso-ciated with return period/level estimates are very high.

Q: Why is uncertainty so important?

19

Page 20: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

A word of caution

Because we use one observation per year, uncertainty asso-ciated with the previous return period estimate is very high.

Q: Why is uncertainty so important?

“There are known knowns; there are things we knowwe know. We also know there are known unknowns;that is to say we know there are some things we do notknow. But there are also unknown unknowns - thereare things we do not know we don’t know.”

Former Secretary of Defense Donald Rumsfeld

Think of uncertainty as a ‘known unknown’. Important toacknowledge that this exists and quantify it.

This is the crux of Mathematical Statistics.

20

Page 21: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

GEV model fit

Plot shows annual maxima 1902-2008.

0 5000 10000 15000 20000 25000 30000

05000

10000

15000

20000

25000

GEV Q-Q plot

Model

Empirical

This model is a better fit. But are we missing something?

21

Page 22: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

GEV model fit

Plot shows annual maxima 1902-2008.

0 5000 10000 15000 20000 25000 30000

05000

10000

15000

20000

25000

GEV Q-Q plot

Model

Empirical

This model is a better fit. But are we missing something?

Non-stationarity? Six of the seven largest flooding eventsoccurred in the last 15 years.

Could be handled by a regression-type approach.

22

Page 23: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why use only ‘extreme’ observations?

Two approaches for extracting extreme observations:1. Block-maximum approach (done above)2. Threshold-exceedance approach (skipped today)

Heuristic explanation: Phenomena which generate extremeobservations are different than those which generate typicalobservations (Red River floods?).

Mathematical explanation: Assume Xt has cdf FX(x).

FMn(x) = P (Mn ≤ x) = P (Xt ≤ x for all t = 1, . . . , n)

= P n(Xt ≤ x)

= F nX(x)

If we know FX exactly, then we know FMn exactly. But ifwe have to estimate FX, any errors in estimating the tail getamplified by a power of n.

23

Page 24: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Generalized Extreme Value distribution

The GEV distribution captures three types of tail behavior.

F (x) = exp

{−[1 + ξ

(x− µσ

)]−1/ξ}.

The parameter ξ determines tail behavior and is difficult toestimate in practice.• ξ < 0: Weibull case (bounded tail)• ξ = 0: Gumbel case (light tail), interpreted as limit• ξ > 0: Frechet case (heavy tail)

Q: Why is the GEV the right distribution to fit to annualmaximum data?

Recall from your introductory statistics course: the CentralLimit Theorem says that the Normal distribution is the rightdistribution for sums of iid random variables...

24

Page 25: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why the GEV?

...there is a ‘CLT-like’ result for block maxima.

Let Mn = maxt=1,...,nXt, where Xt are iid. If there exist nor-

malizing sequences an and bn such that P(Mn−bnan≤ x

)→ G(x)

(nondegenerate) as n→∞, then

G(x) = exp{− [1 + ξx]−1/ξ

}.

We don’t need information about the distribution of Xt toknow about the distribution of Mn.

25

Page 26: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Limiting Distributions

-2 0 2 4 6

0.0

0.1

0.2

0.3

0.4

0.5

GEV distributions

density

Weibull (ξ = −0.5)Gumbel

Frechet (ξ = 0.5)

26

Page 27: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Bivariate Extremes

Assume Xt = (Xt,1, Xt,2) iid.

Idea: Describe the joint tail of the distribution

Goal: Often extrapolation

0 500 1000 1500 2000 2500

010000

20000

30000

Daily Footprint from WRFG and Gridded Obs, Original Scale

WRFG Output

Obs

erve

d P

reci

p

2/17/86

1/9/95

12/12/95

11/19/96

1/1/97

27

Page 28: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Bivariate Extremes: Example

Portfolio consisting of two securities. What is the probabilitythat one goes bust, given that the other does?

The Gaussian copula was a popular model for portfolio risksat least until 2008:

28

Page 29: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Bivariate Extremes: Example

Portfolio consisting of two securities. What is the probabilitythat one goes bust, given that the other does?

The Gaussian copula was a popular model for portfolio risksat least until 2008:

Wired magazine: “Recipe for Disaster: The Formula that Killed Wall Street”. February 23, 2009.

Basic problem: this model doesn’t account for multiple se-curities experiencing an ‘extreme event’ at the same time.

Some forward-thinking statisticians in extreme value theorywere warning of this problem years ahead of time.

Their warnings went largely unheeded.

29

Page 30: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Bivariate Extremes

Marginal and dependence effects are typically handled sepa-rately.

Approach:1. Transform marginal to something nice.2. Describe dependence.

●●●

● ●

●●● ● ● ●

●●●

●●

●●●●

●●

●●

●●

●●

●●

●●

●●●

●●

● ●●

●●●

●●●

●●

●●

●●

●●

●●

●●

●●

●●●

●●●

● ●

●●

●●

●●

●●

● ● ● ● ● ●●

●●●

●●

●● ●

● ● ●

●●

● ●●●●

●● ●

●●●●

●●

●●

●● ● ●

●●●

●●●

●●●

●●

● ●●●

●●

●●

●●

●●●●

● ●

●●

●●

●●

●●

● ● ●●

● ●

●●●

●●●

●●●

● ● ● ● ● ●

●●

●●●●●●●

●●

●●

●●

● ●●●●

●● ● ●

●●

●●

●●

●●

●●●●

●●●●●●

● ● ●

●●●

●●●

●● ● ●

●●

●●● ●

●●

●●

●●● ●

●●

●●

●●

●●●●●

●●●

● ●

●●

●●

● ●

●●

●●●

●●

●●

●●●

●●●●

●●

●●

●●● ●●●

●●

● ●

●● ● ● ● ●

● ● ● ● ●●

●● ●●●

●●●

●●●●

●●

●●●

●●

● ● ●●

● ●●

●●

●●

●●

●●

●●●

● ●●

●●

●●

●●●●

●●

● ●

●●●

●●

●●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

● ● ● ●●●●

●●

●● ● ● ●

●●

●●●

●●

●●

● ●●●●

●●

●●

●●

● ● ●●

●●● ●

●● ●●

●●

●● ●

● ●●

●●

●●

●●

●● ● ●

● ●●●●●

●●●●

●●

● ●●●●●

●●

●●

●●

●● ● ● ● ●

●●●●

●●

●●

●●

● ●●

●●

●●

●●

●●●

●●●

●●●●●

●●

●●

●●

●●

● ●●

●●●●

●● ● ●

●●●●●

●●●●●

●●

●●

●●●

●●● ● ●

●●

●●●

● ●

●●

●●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●

●●

●●● ●

● ●● ●●●

●●●

●●●

●● ● ●

● ●

●●

● ●●●●

● ● ● ●●●●

●●

●●

●●

●●

● ●

●●●

●● ●

●●●●●

●●

●●

●●

●●

●●

●●

●●

●●●●

●● ●●●

● ●

● ●●

●●

●●

●●●●●

● ●

●●

●●

●● ●●●

●●

●●

●●●

●●●

●●●

● ●●

●●●

●●●●●

● ●

●● ●

●●

●●

●●

●●

●● ● ●

●●

●●

●●

●●

●●

●●

●●

●●●●

● ● ●●●

●●

●●●

● ●●

●●

●●

● ●

● ●●●●●

●●

●●

●●●

●●

● ●●●●●● ●● ●●●●●●●

●●

● ●● ●

●●

●●●●

●●●●

●●

●●

●●●●

● ●●

●●●

●●

● ● ● ●●●

●●

●●●●●●

● ● ● ●●●

●●

●●●

● ●●

●●●

●●●● ●●

●●

●●

●●

●●●

●●

● ● ●●●

●●

●●●

●●●●

●● ●

●●●

●● ●●

● ●●

●●●●

●●

● ●●

●●

●●

● ● ●●

●●

●●

●●● ●

●●

● ●

● ●●

●●

●●●

●●

●●

●●

● ●●

●●●●

●●●●

●●

● ●

●●●

●●●●

●●●

●●

● ●● ●

●●

●●

●●

●● ● ● ● ●●

●●

●●

●●

●●

●●

●●

●●

● ● ● ●●

●●

●●●●●

●●

●●

●●

●●●

●●

● ● ● ●●

●●

●●

●●

●●

● ● ●● ● ●

●●

●●●

●●

●●●

●●

● ●

●●●

●●

●●

●●●●

●● ●

●●

● ●●●●●●●

●●

●● ●

●●

●●●

●●

● ●●

● ●

●●●

●●

●●

● ●●

● ● ●● ●

●●●

●●●●●●

● ●

●●

●●●

●●● ●

●●

● ●

●● ● ● ●

●●

● ●●●

●●●●

● ●●● ● ●

●●●

●●

●●

●●●

●●●●

●●●

●● ●

●●

●●

●●

●●

●●●

●●●

●● ●

●●●

●●

●●

●●●●●

●●

●●

●●● ● ● ●●

●● ● ●●●●

● ●●●

● ●●●●●

●●● ● ● ●●

●●●●●

●●

●●●●●●●

●●●●●●

●●

●●

●●●●●

●●

●●

●●

● ● ●●

●●●●●●●●●

●●

●●●

●●

●● ●

●●●

●●

●●●

●●●

● ●● ●●●●●●●●●● ● ● ●●● ● ●●●

● ● ● ●●●

●●●●●●

● ●● ●

●●●●●●●●

●●●●●●●● ●

●●●●●●

●●●

●●●

●●●●●●

● ●●

●●●●

●●

●●●● ●

●●●●● ● ●

●●●●●

●●

●●

●●

●●

●●

●●● ● ●

●●●

●●●

●●●●

●●● ●● ●●●●

●●●

● ●● ●

●●●

●●●●

●●●● ● ● ●

●●●●●●●●●●●●●●●●●

●●●●●●

●●●●

●●●●●●

●● ● ● ●●

●●●●

●●●

●●●●●

●●

●●

●●●●

●●●●

●●

●●●●●●●

● ● ● ●

●●●●

●●

●● ●● ●● ●●

●●

● ●●

●●

●●●●●

●●●●●

●● ● ●●

●●●●●●●●●●

●●● ●●●●●●

●●●

● ●●●●

●●

●● ● ●●

●●●●

●●●

● ●●●

●●

●● ●●●

●●●

●●

●●

●●●●●

● ● ● ●●●●●●●●

● ●●●●● ●●●●● ●●●●● ●●●● ●●● ●●

●●

●●●

●●

●●

●●●● ●●●

●●

●●●

●●

●●●●

● ●

●●

●●● ●●

●●

●●

●●●

●● ● ●●

●●●●

●●

●●

●●●●●●

●●●●

●●

●● ●

●●

●●●●

●● ● ●●●●

●●

●●●

●● ●

●●●

●●

●●

●●

●●

●●●●●●

●●

●●

●●

●●

●● ●●

●●●

●●

●●●●●

●●

●●●

●●●●●●●●

● ● ●

●●●

●●

●●●●

●● ●

●●

●● ●

●●

●●

●●

●●

●●

●●●

●●●

●●

●● ●

●●

●●

●●

● ● ●

●●

●●●●

●●

●●●

●●●

●●●

●●

●●

●●●

●●●

●●

●●

●●

●●●●

●●●

●●●

●●

●●

● ●●●●●

●● ●

●●●

●●

● ●●●●●●

●●

●●

●●●

●●●●

●●●

●●

●●

●●

●●

●●

●●

●●●●●

●●●

●●●●

●● ●

● ●●●●●

●●●

●●●

●● ● ●●●●

●●●●●●●

●●

● ●●●●●

●●

●●●

●●●●

●●●●

●●

●●●

●●

●●●●●●

●●●●●

●●●●●

●●

●●

●●

●●

●●

●●

●●

●●

● ● ●

●●●●●●●

●● ●

●●●

●●

●●

● ●●●●●

●●

●●

● ●●

●●

●●

●●

●●

●●

●●●

●●

● ●

●●

●●●

●●

●●

●● ●

●●

●●●●●●●

●●

●●

●●●

●●

●●

● ●●

● ●●●

●●

●●

●●

●●●

●● ●

●●

●●

●●

●●

●●

● ●● ●

●●●

●●

●●

●●●

●●

●●

● ● ●●●●●

●●

●●

●●●

●● ●

●●

●●●●●●

●●●●●●●

●●

●●●●●

●●

●●

●●●

●●

●●●

●●

●●●●

●●●●●●

●●●

●●●

●●

●●●

●●

●●●●●

●●

●●

●●●

●●

●●

●●●●●

●●

● ● ● ●●●

●●●

●●●

●●

●● ●

● ●●

●●

●●●

●●

●●●

● ●● ●●

●●●

●●

●●

●●

●●

●●

● ●●

●●

●●

●●●●●●

●●

●●

●● ●

●●●

●●

●●●

●●

●●

●●

●●●●●●

●●

●●

● ● ●●

●●●

●●●

●● ●

●●

●●●

●●●

● ● ●● ● ●

●●●●

●●

● ●●

●●●● ●

●●●●

●●

●●

● ●

●●

●● ●

●●

●●

●●

● ●

●●●

●●●●●

●●

● ●

●●

● ● ●

●●

●●● ●

● ●

● ●

●●●

●●

● ●●

●● ●

● ●●

●●

● ●●

●●

●●

●●

●●

●●

●●●

●●●

●●

●●

●●

● ●

● ● ●●

● ●

●●

●●

●●

●●●●●●

●●

●●

●●

●●●

●●

● ●

●●

●●●●

●●●

●●

●●

● ● ●●●●

●●

●●

●●

●● ●

●●●●

●●●

●●

●●

● ●

●●●●

●●

●●

●●

● ●

●●●●

●● ●

●● ●

●●

●●

●●

●●

●●

●●●

●●

●●●●

●● ●

● ● ●●●

●●

●●

●●●

●● ●

●●●●●

●●●

●●●

● ●

●●

●●

●●

●●

● ●

●●

●●

●● ●●

●●●●●●●

●● ● ●

● ●●

● ●●●

●●●

●●

● ● ●●

● ●●

●●●

●●

●● ● ●

●●●

●●

● ●

●●

●●

●●

●●

●●●●●

● ● ●●

●●●

●●●

●●

●● ● ●●

●●

●●

●●●

●●●●

●●

● ●

● ● ●●

●●

●●●●●

●●

●●

●●

●●

●●●

●● ● ● ● ●

●●

●●

●●

●●●●●

●●

●●●

●●

●●

●●

●●

●●

●●

●●●●

●●

● ●

●●●●●●●

● ●●

● ●

●●●

●●

●●●

●●●

●●

●●

●●●

●●●●

●●

● ●● ● ●

●●

●●●● ●●

●●●

●●

● ●●●●●

●●●●●●

●●●

● ●

●● ●

●● ●●

●●

●●● ●

●●●● ●

●●●

●●●

●●●●

● ● ●● ●

●●●

●●●●●●●

● ● ●

●●

●●●●

●●●●

●●

●● ●

●● ●

●●●

●●●

●●

●● ●

● ●●●●

●●

●●●●●●

●●

●●

●●●●●●

●●●

●●

●●

●●

●●

●●●●●

●● ● ● ●

●●●●

●●●●●●

●● ● ●●

●●●

●●

● ●

●● ●

● ● ●●●

●●●

●●

● ●● ●●●

●●●

●●●●

● ●● ●

●●

●●

●●

●●

●●

●●

● ●

●● ●●

● ●

●●●●●●

●●

● ●● ●

● ●●

● ●●●●

●● ● ●

● ●●

●●●

●●●

●●●●

●● ●

●●

●●

●●

●●●●

● ●●

●●●●

●●

●●

●●●

●●

●● ● ●

●●●

●●

●●●

●●●

●●

●●● ●

●●●

●●●●●

●●

●●

●●

● ●●

●●

●●●

●●

●●

● ● ●●

●●●●

●●●

●●●●

●●

●● ● ●

●●●●

●●

●●

●●●

● ●●

●●●●●●

●●

●●

● ● ● ●●

●●●

●●●●●●●

●● ●

● ●●

●●●

●●●

●●●

●●

● ● ● ●●●●

● ●●●●

●● ●

●● ●

● ●●

●●

●●

●●

●●●●

●●

● ●● ● ●

●●

●●●

●●● ●●●●●

●●●

●●

●●

●●

●●

●●

●●

●●●●

●● ●

●●●●

●●●●

●●

●●

●● ●

●●●●●●

●●

●●

●●

●●

●●

●●

●●●

●●

●● ● ●

● ● ●●

●●●

●●

●●

●● ●

●● ● ●●

●●

●●

●●●

●●

●●

●● ●

●●

●●

●● ●

●●

● ●●

●●

●●

●●

●●

●●

●●

●●

●●

●● ● ● ● ●●●

●●●

●●

●●

●● ●

● ●

●●

●●

●●

●●●

●●

●●

● ●

●●●

● ●● ●

●●

●●●

●●●

●●●

● ●●

●●

●●●

●●

● ●●

● ●

●● ●●●

●●

● ●●●●

●●

●●

● ●●

●●●●●

●●●

●●

● ●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●● ●

●●

●●

●● ●

●●●

●●

● ●

●●●

●●

●●●●

●●

●●●

●●●●●

●●

●●

●●

● ● ●●

●●●

●●●

●●●

●●

●●

●●

●●●

●●

●●

●●●

●●

●●

●●

●●

● ●

● ●

●●

●●

●●●

●●

●●

●●

●●

●●●●●●

● ● ●● ● ● ●

●●

●●●●●●

● ● ● ● ●●●

●●

●●

●●

● ● ● ● ● ●●●●

●●

●●

●● ● ● ●

●●

●●

●●●

● ●

●●

● ●

●●

●●

●●

●●

●●

●●

●●

●●●

●●●

●●●

●●

●●

● ●●●●●

●●●

●●

●●

●●

● ●● ●

● ●●

●●●●

●●●

●●

● ● ● ●●●

●●●

●● ● ●

●●

●●●

●●

●●

●●● ●

●●●● ● ●

●●

● ●●●

●●●

●●

●●

●● ●

●●●●

●●

●●

●●●

●●●

●●

●●●●●

●●

● ● ●

●●

●●

●●

●●

●●

● ● ●●●

●●●●● ●●

●●

●●

●● ●●

●●

●● ●

● ●

●●

●● ●

●●●

●●●

●●

●●

● ●●

●●●●

●●

● ●●

●●●●

●●

●●●●●●●

● ●●

●●

●●

●●

●●

●●

●●

● ● ● ●●●

●●●●

● ●●

●●

●● ●

●●

●●●●

●●

●●

●●●

●●●

●●●

●●●●

●● ●●

●●●●

●●●●●●●

●● ●

●●

●●

●●●

●●

●●

●● ●

●●

●●●

●●●

●●●

● ●●

● ●●

●●●

●●

●●●

●●

●●●

●●

●●

●●

● ●●

●●

●●

●●●

●●●

●●

● ● ●●●

●●●

●●●●●●●

●● ●●

●●●●

●● ●

●●

● ● ●●●●●

●●●

●●

●●

● ●●

●●

● ●●

●●

●●

●●

●●●

●●

●●

●●● ●

●● ●

●●

●●

● ●●

●●●

●●●●

●● ● ● ●

●●●

●●

●●●●●

●●

●●

●●●●● ●

●●

●●

●●

● ●

● ●●

●●

●●

●●

●●

● ● ●

●●●●

●●●●

●●●

● ●●

●●●●

●●

●●●

●● ● ●●

●●

●●

●●

●●●

● ●● ● ●

●●●

●●

●●

●●

●●

●● ●

●●

●●●●

●●

● ●● ●

●●

● ●●●●●

●●

●●●●

●●

●●●●●

● ●

●●●

●●●

●●●●●

● ●●●

●●

●● ●●

●●

●● ●

●●

●●●

●●

● ●

● ●

●●

●●

●●

●●

● ● ● ●●●

●●●

●●

●●

●●

●●

●●●●●

● ● ●

● ●

●●●

●●

●● ●

●●

●●●●

●●●

●●

● ● ● ● ● ●●●●

●●

●●

●●●●●

●●

●●

●●●

●●●

●●

● ● ●

●●●●

●●●●

●●

●●●

● ●●

● ●

●●●

●●

●●

●●

●●● ●

●●●

● ●

● ●●

●●●

●●●●●

● ● ● ● ● ●●●●

●●

●●●

● ●●

●●● ●

● ●● ●●

●●

●●

●●

●●

●● ● ●

●●●

●● ●●

● ●●

●●

● ●

●●

●●

●●

●●●

●●●

● ● ●●

●●●

●●

●● ●●●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●●

●●●●

●●●●

● ●

●●●●

●●

●●

●●●

●●●●

●●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●●

●●●

●●●●

●●●●

●●

●●

●●●

●●

●● ●

●●

●●

●●

● ●● ●●

●●●●

●●●

● ● ● ● ●

●●

●●

●●

● ● ● ●●● ●

●●●●

●●

●●

●●

●● ●

●●●●

●●

●●

● ●

●● ●●

●●●

●●●●●

●●●

●●

●●●

●●●●

●●

●●

●●●

●●

●●●●

●●

● ●●●●

●●

●●● ●●●

●●

●●

●●

●●

●●●

●●●

●●

●●

●● ●●●●

●●●●●

● ●●●●●●

●●●●

●●●

● ●●

●●●

●●

●●

●●●

●●●

●●

● ●●●●

●●

●●●

●●

●●

● ●●

●●●

●●

●●●●●●●

●● ●

● ●

●● ● ●●●

●●

●● ●

●●●●

●●●

●● ●

● ●●●●●

●●●

●●●

●●●●

●●●●

●●●●

●●

●●●

●●

●●●●

●●

●●

●●

●●●

●●●

●●●●

● ●

●●●

●●

●●

●●

●● ●

●●

●●●●●

●●●

●●●●●

●●

●● ● ●

●●

●●●

●●●

●● ●

●● ●

●●

●●

●●●

●●

●●●●

● ●●●●●●

● ●●●●●

●●

●●●●●

●●

●●

●●●●●

●●

●●

●●●

●●●

●●●●●●●

●●

●●

●●●●

● ● ●● ●

●●●

●●

●●

●●

● ●●

● ● ●●●

●●●

●●

●●●

● ●●

●●●●

●●

●●

● ● ●●

●●●●● ●●

●● ●

●●●

●●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●● ●●

● ● ●

●●

●●●

●●

● ●● ● ● ●

●●●

● ●

●●

●●

●●

●●

●●

●● ● ●

●●

●●

●●

●●

● ●

●●

●●

●●●●●

●●

●●

●●

●●●●

●●

●●●

●●●

●●

● ●

● ●●●

●●

●●●●●●●

●●●

●●

●●●

●●●●●●●

●●

●●

●●●

●●

●●●

●●

●●

●●

●●

●●

●●

●●

● ●

● ● ●●

●●

●●●●●●

●●●

●●

●● ●

●●●●

●●

●●

● ●

●●●●●

●●●●●

●● ● ● ● ●●

●●●●●●

●● ●

●●

●●

●●●

●●

●●

●●●

●●●●

●●

● ●●

●●

●●

●●

●●

●●●

●●

●●

●●●●●●

●●

●● ●

●● ●

●●●●

●●

●●

● ●●

● ●

●●●●●●

● ●●

●●

● ●●

●●

●●

●● ●

● ●● ●

●●●

●●●

● ●●

●●

●●●●●●

●●

●●

● ●●●●

●●

●●

●●

●●● ●

●●

●●

●●●

●●●●

●●●●●●

●●

●●●

●●●

●● ●

●●●●

●●

●●●●

●●●

●●

●●

●●

●●●●

●●●●

●●

●●●●●

●●

●●●

●●●●

●● ●

●●

●●●●●●●

●●

●● ●

●●

●●● ● ●●●●

●●

●●

●●

● ●●● ● ●●●

●● ●

●●

●● ●●●●

●●

● ●●

●●

●●●●

●●

●●●●

●●

●●

● ●●●●●

●●

● ●●●

●●●●

● ●●●

●●

●●

●●●●

●●

●●●●●●

●●

●●●●●

●●

●● ● ●●

●●

●●●

●●

● ●●●

● ●●

●●

●●●●

●●

● ●● ●●

● ●●●●

●●

●●●

●●

● ● ●● ●

●●

●●

●●●

● ●

●●

●●

●●

●●

●●

●●

●●

●●●●●●

●●●●●● ●

● ●●●

●●●●

● ● ●●

●●

●●

●●

●●

●●

●●

● ●●●

●●

●●●

●●

●●

● ●

●●

● ●●

●●●●●●

●● ●●

●●●●

●●

●●

●●

●●

● ● ● ● ●●●●

●●

● ● ● ●●

● ●●

●●

● ●

●●

●●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●

●●●●

●●●

●●

●●

● ●●

●●●●●●

● ● ●

●●●

●●

●●●

●● ●

●●

●●●●

●●

●● ●●

●●●

●●

●●

● ● ●●

●●●●

●●● ●

●●●

●●

●●

● ●●●●●

●●●

●●●●

● ● ●●●

●●

●●●●●●●●

●● ● ●

●●

●●

●●●●● ●

●●

●●

●●●●●●●

●●

●●

●●●

● ●● ●●

●●

●●

●●●

●●●

●● ● ● ●●

●●

●●

●●

●● ●

●●

●●●●

●●

●●●

●●

●●

●● ●●●●

●●●●●

●●

● ● ●●●●

●●●●●●●

●●

●● ●

●●

●●

●●●

●●

●●

●●

● ●●

●●

●●●

●●●

●● ●●

●● ●

●●

●●

●●

●●●

● ● ● ● ●●●

●●●●●●

●●

●● ●

●●

●●●●●●

●●

●●

●●

●●●●●●

●●

●●

●●●●●●

●●

●●

●●●

●●●

● ●●●

●●●

●● ●

●●

● ● ● ●●●●

●●

●●●●●

●●

● ● ● ●●●

●●

●●●

●●

●● ●

●●

● ● ●●●●

●●

●●

●●●●

●●

●●●●

● ●

●●

●●●

● ●● ● ●●

●●

● ●

●●●

●●

●●

●●

●●●●

●●●●

●● ● ●

●●

●●●

●●●●

●●

●●●●

●●●

●●

●●●

●●

●●●●

●●

● ●●●

●●●●

●●● ●

●●

●● ●

●●●●

●●●●

●●

●●●●

●●

●●●●●●●

●● ● ●

●●●●

●●●●●●●●

●●

●●

●●●●●●●●●

●●

●●●

●●

●●

●●

●●

●● ●

● ●●

●●

●●

●● ●●

●●●

●●

●● ●

●●

●●

●●

●●

● ●●

●●

● ●●

● ● ●

●●

●●

●●●

●●

● ●●●

●●●

●●●

●●

●●

● ●●

●●●●●●

●●

●●

●●

●●

●●●●●●

●●

●●

●●

●●●●

●●●●●●

●●

● ●

●●

●●

● ●● ● ● ●

●●●

●●●●●

●●

●● ●

●● ●●●●

●●

●●

●●●●

●●

●●●●

●●

●●●

●●●

●●

●●

● ●●●●

●●

●●

●●

●●

●●

●●●

●● ●

●●●

●●

●●●●●●

●●

● ●●●●●

●●●●

●●

●●

●●

●●●

●●●●

●●●

●●

●●

●●

●●

●●●●●

● ●● ●

●●

●●●

●●● ●

●●

●● ●●

●●●

●●●

●● ●

●●

●●●

●●●

●●

●●

●●

●●

●●●

●●●

●●

●●

●●●

●●

●●●

●●●

● ●●

● ●

●●●

●●●●

●● ●

●●

●● ● ●●●●

●● ●●

●●

●●

●●●●●●

● ●●

●●

●●●●●●●

●●

●● ●●●

●●

●●●●●●●

● ●

●●●●

●●●●●●●

●●●

●●●

●●●●

●●●●

●●●●

●●

● ●●

● ●● ●●●

●●

●●

●●

● ● ● ● ●●

●●●●

● ●● ●●●●●●●

●●

●●

●●

●●

●●

●●●

● ●● ●

●●

●●

●●

●●

● ●●●

●●

●●

●●●●

●●

●●●

●●●●●●●

●●

●●

●●

●●

●●

●●●

● ●●

●●

●●●●

●●●●●●●

●●

●●

●●

● ●●

●●

●●

●●

●●●●

●●●●

● ● ● ● ● ●●●●

●●

●●

● ● ●●●●●●●●

●●

●●●●●

● ●● ● ●

●●

●●

●●

●●

●●●

●●

●●●

●●

●●

●●

●●●●●

●●●●

●●●

● ●●

●● ●

●● ●

● ●

●●●

●●

●●

●●

●● ● ●

●●

● ●●

●●●

●●●●

●●●●●

●●

●●●● ●

●●●

●●●●

●●

●●●●●

●●●

●●

●● ●

●●

●●

●●

●●●

●●

●●

●●

●● ●

● ●●●●

●●● ● ●●

●●●●●●●

●●●

●●●●●●●

●●

●●

●● ●

●●●●● ●

●● ●

●●●●

●●

●●●●●●

●●

●●●

●●

●● ● ● ●●

●●

●●

●●

●●

●●

●●●

●●●●●●●

●●

●●●

●●●

●● ● ●

●●●●●●

●●●

●●

●●

●●

●●

●●

●●

● ●●●

● ●●●●

●●●●●

● ●

●●

●●

●●● ●

●●●●

● ●●●●●

●●

●●●●

●●●

●● ●

● ●●

●●

●●

●●●●●●

●●●

●●

● ●●

●● ●●

●●●

● ●●

●●●●

● ●

●●

●●

●●

●●●

●● ●

●● ●●●● ●

●●●●●

●●●●

●●

●●●

●●●

●●● ● ●

●●

●●

●●●●●●●●●●●●

●●●●●

●●

●●●●

●●●

●●●●

●●●

●●

●●

●● ●

●●●●●

●●●

●●

●● ●

●●

● ●●

●●●

●●●●

●●

● ● ● ● ●●

●●

●●●● ●

●●

●● ●

●●

●●●●●

●●●

●● ●

●● ●

●●

●●

●●●●●●●●

●● ●

● ● ●●●●●

●● ● ●●●●●● ● ●●●●●●●● ●● ●●●● ●● ●●●●●● ●●●●●●●●●● ●●●●●●●●●● ●●

●●●●

●●

●●

●●

●●

●●

●●●●●

●●

●●

● ●●

● ●● ●●

●●●

●●●●

●●●

●●

●● ●

●●

●●●●●●

●●● ●

●●

●●

●●●●

●●●

●●

●●

●●

●●

●●●

●● ● ● ●● ●

●●●● ●

●●

●●●●●

●●

● ● ●●●

●●●●●

●●

●● ●●

●●

●●

●●

●●●

●●●

● ● ●●

●● ●●●

●●●●●

●●

●●●●●

●●●●

●●

●● ●●

●●●

●●

●●●●

● ●●

●●

●● ●●●●●

●●

●●●●●●●●

● ● ●●●●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●●●●●

●●

●●●

●●●

●●●●●●●●●●●●

●● ●

● ●● ●

●●●●

●●●●●●

●●

●●

●●●

●●

●●

●●

●●●

●●

● ●●●

●●

●●

●●

●●●

●●

●●●●●●

●●●●

●● ●

●● ●

●●

●●

●●

●●

● ●

●●

●●

●●

●●

●●

● ●●

●●

●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●

●●

●●

●●●●

● ● ● ●●

●●●●

●●

●●●●

●●

●●

●●●

●●●

●●●

●●

●●

●●

●●

●●

●●●

● ●

● ●●●

●●●●● ●

●●●

●●

●● ●●

● ● ●●●

●●

●●

●●

●●

●●

● ●●●●●●●

●●●●●●●

● ● ●

●●●●

●●

●●●●●●●●

●●●

●●

● ●

●● ● ●●●

●●●●

●●●●●●

●● ●●

●●●●

●●

●●

● ●●●●

●●●

●●

●●

●●

●●●

●●●● ●●●●

●●●

●●

●●

●●●

●●●

●●

●●●

● ●●●

●●●

●●

●●●●●●●

●●●

●●

●●

●●

●● ●●

●●●

●●

●●●

●●

●●

●●

●●●●

●●

●●

●●●●

●●

●●

●●●

●●●● ●●

● ●

●●

●●

●●●

●●

● ●●●●●●●●

●●

●●●

● ●

●●

●●

●●

●●●●

●●

● ●●●●

●●●

●●●

●●●

●●

●●

●●●

●●

●●●●

●●●

●●

● ●●●

●●●

●●

●●●●

●●

●●●

●●●

●●●

●●●●

●●

●●

●●●●●●

●●

●●●●

●●

●●

● ● ●

●●

●●

●●

●●●

●●

● ●

●●●●

●●

●●

●●

●●

●●

●● ●●

●● ● ●●●●●●● ●●●●●●● ●●●●

●●●●●●

● ●●

● ●●

●● ●

●●●●●

●●

●●

● ●●

●● ● ●●●

●●●

●●●

●●

●●

●●●

●●

●●

● ●●

●●●

●●

● ● ● ●●

●●●●

● ● ●●

● ● ● ●●●

●●●●●

●●●●●●●●●

●●

●●●●●●

●●

●●

●●●●

●●

●●

●●

● ● ●●● ● ●

●● ●

●●●

●●●

●●

●●●● ●

●●

●●

●●

● ●

●●●

●●

●●●● ●

●● ● ●●●

●●●

●●●

●●

●●

●●

● ● ●●

●●

●●

●●

●●

●●

●●

●●

● ●

●●

●●

●●●

●●

●●

●●

●● ●

● ●●

●●●

●●

●●●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●●●

●●

● ●

●●

●● ●

●●●

●●●

●●●

●● ●

●●

●●●

●●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●●

●●●●

●●

●●●

●●

●●

● ● ●

● ●

●●●

● ● ●●

●●

● ● ● ●●●

●●

●●●

●●●

●●

●●

●●

●● ●

●●

●●●

●●

● ●

●●

●●●

●●

●●

● ●●●

●●●

●●●●

●●

●●

●●

●●●

●●

●●

●●

● ●● ●

●●●

●●

●●●●●

●●

●●

● ●

●●

●●

●●

●●

● ●

●●●

●●●● ●

●● ●●●

●●

●●

●●●

●●●

●●●

●●●

●●

●●

● ● ●●

●●●●

●●●●●●

● ●●

●●

●●

●●

● ● ●

●●●●●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●●

●●

●●

●●

● ●

●●

●●●

●●

● ● ●●●

●●

●●●●●●●

●●

● ●●

●●●

● ●●

● ●●

●●●

●●

● ●

●●

●●

●●

●●●●●●

● ● ●

●●

●●

●●●

●●●

● ●●

●●

● ●●●

●● ● ●●●●●

●●

●●●●●

●● ●

● ●●●

●●

●●

●●

●●

●●●●

●●

●●

●●

●●

●●

●●

●●●●●●

●●●

●●

●●

●●●

●●

●●

●● ●

●●

●●● ● ●●

●●●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

● ●●

●●

●●●

●●

●●

●● ●

●●

●●

●●

●●●●

●●●

●●

● ●

●●

●●

●●●

●●

● ● ●● ●●

●●

●●

●●●

●●

●● ●

●●●●

●●

●●

●● ●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●●●●●●●

●●

●●●●

●●

●●

● ●●

●●●●

●●

●●

●●

●●●

●●●

●●

●●

●●

●●

●●

●● ● ● ●

●●

●●

●●

●●

●●

●●

●●

●●●●

●●●●

●●

●●

●●

●●

●●●●●

●●●● ●●

●●●

●●●●●●●●

●● ●

●●

●●

●●●

●● ●

● ● ●●

●●

●●

●●

●● ●

●●●

● ●●●●●●●

●●

●●●

● ●

●●

● ●●●●●

●●

●●

● ●●

●●●

●●

●●

●●

●●●

●●

● ●

●●●

●●

●●

●●

●● ●●

●●

●●

●●

●●

●●●●

●●●●

●●●

●●

● ●

●●●●●

●●

●●●●

●●●

●●

●●

●●●

●●

●●●●

●●●●●●●●

●●●●●●●●●

●● ● ●

●●

●●

●●

●●

● ●

●●

●●●

● ●

●●● ●●●

● ●●●

● ● ●●

●●

●●●

●●●

●●●●

●●

●● ●●

●●●

●●

●●

●●

●●

●●●

●●

●●●●

●●● ●● ●● ● ●

●●

●●

●●

●●

● ●

●● ●

●●

●●

●●●●

● ●●

●● ●●

●●

●●●●●

●●

●●

●●

●●●

●●

● ●

●●●●

●●

●●●

●●

● ●●●●●

●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●●●

●●●●●●●

●●

●●

●●

●●●● ● ●

● ●●

●●

●●

●●

●●●●

●●

●●

●●●●●

●●●

● ●●●●

●●

●●●

●●

●●

●●●

●●

●●●

●●

●● ●

●●

●●●●

● ●

●●

●●●

●●●

● ●

●●●

●●

●●

● ●●

● ●

● ●●●

●●

● ●

●●●

●●

●● ●

●●

●●●

●● ●

●●●●

●●

● ●●

● ●

●●●

●●

●●

●●●

●●●

●●

●●●●●

●●●●●●

●●●

●●

●●●

●●

●●

●●●●●●

●●●●●●

●●

●●

●●●●●

●● ●●

●●●

●●

●●

● ● ●● ●

●●●●●●

● ●●●

●●

●● ●

● ●●

●●●

●●

●●●

● ●●●●●●●●●

●●

●●●

●●●

●●

●●●

●●●

●●●●●●

●●

●●●

●●●●●

●●●●●●●

●●

●●●● ●

●●

●● ● ● ●●●

●●●●●●

●●●

●●●●●●●

● ●

●●● ●●●●

●●

●●●●

●●

●●●

●●●

● ●●

●●●

● ● ●

● ●● ●

●● ● ●

●●●

● ●●

●●●

●●●

●●

●●

●●

●●●

●●

●●●●●●

●●●●

●●●●

●●

●●

●●●●

●●

●●●

●●●

● ●●

●●

●●●

●●●●

●●

●●

● ● ●●●

●●●

●●●●

●●

●● ●

● ●●

●●

●●

●●●

●●●●●

●●●●●

●●

●●

●●●

●●

● ●●●●

●●●●

● ● ●●

●●

●●

●●●●●

●●●

●●

●●

●●●

●●

● ●●

●●

●●

●●●

●●

●●●●

●● ●

● ●●

●●

● ● ● ●●

●●

●●●

●●●

●● ● ●●●●●

●●●●●●

●● ●

●●●●●●●●

●●

●●●●

●●

●●

●● ●

●●●

●●●

●●

●●

●●

●●

●●●●

●●

●●

● ●●

●●●●

●●●

●●

●●●

● ●●

●● ●

●●

●●●

●●

●●●

●●

●●

●● ● ●

●●

●●●

●●●●●

● ●●

●●

●●●

●●

●●●●●

●● ● ● ●

●●

●● ● ●

●●

●●

● ●

●●

● ●

●●●

●●

●●

●●●●

●●

●●●

●●

●●

●●●●●●

●●

● ●●●●

●●

●●

●●●● ●

● ●●

●●

●● ●

●● ●

●●●

●●●

●●●●

●●

●●

●●

●●

●●

●●●

●●

●●

● ●● ●

●●

●●

●●

●●

●●

●●

●●●

● ●●

●●●●

●●

●●

●●

●●●●●●●

● ●

●●

●●

●●●

●●

●●

●●

●●●

●●●

● ●

●●

● ●●

●●

●●●

●●

●●●

●●

●●

●●

●●

●●

●● ●

●●

●●●

●●

●● ● ●

●●

●● ●

● ● ●

● ●●

●●●●●●●

●●

●●

●●

●●

●●●

●●●

●●

●●

●●●●

●●●

●●

●●

●●● ●

●●

●●

●●

●●●

●● ●●●●●●

●●●●●●●

●● ●●

●●

●●●

●●●

●●

●●

●●

●●

●●●●●●

●●

●●●●

● ●●●●●●

●● ●

●●

●●

●●

●●

●● ●

● ●

●●●

●●

●● ●

● ● ●●●

●●●●

●●

●●

●●●

● ●●●

●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●●

●●

● ●●●

●●

●●

●●

●●●

●●●

●●●

●●

●●

●●

● ●

●●●●

●●●

●●●

●●●

●● ●●●●

●●●

●●●●●● ●●

●●●●●

● ●

●●

●●●

●●

●●

●●

●●

●●

●● ●

●●

●●●●

●●

●●●

●●

●●

●●

●●

●●

●●

●●●●

●●

●●

●●

● ●●

●●

●●

●●●●

●●●●

●●●

●●● ●

●● ●●●

●●●●●

● ●●

●●

●●

●●●

●●●

●●●

●●

●●

●● ● ●

●●

● ●●

●●

●●

●●

●●

● ●

●●

●●

●●

●●

●●●●

●●

●●

●●●

●●

●●

●●

● ● ●●●●

●●●

●●●●● ●●●

● ● ●

●●●

●●

● ● ● ● ● ●●

●●

●●

●●●●

●●●

● ●●●

●●

●●

● ●●●

●●

●●

●●

● ●

●●

●● ●

●●

●●●●

●●

●●●

●●

●●●●

●●

●● ●

●●

●●

●●●●●●

●●

●● ●

●●

●●

●●●●

●●

●● ● ●●●

●●●

●●

●●●

●● ●

●●●

●●

●●●

●●

●●

●●●●

●●●

●●

●●

●●●●

●●

●●●●●

●● ●

●●

●● ●●

●●

●●

●●●

●●

● ●●

● ● ●●●

●●●

●●●●●●

● ●●

●●●

● ● ● ●●●

●●

● ●

●●

●●

●●●●●●

●●

●●

● ●●●●●

●●

●●

●●

●●●

●●●

●●

●●

●●●

●●

●●

●●●

● ●●

●●●

●●

●●●●●

●● ●●●●

●●●●●

●●●

●●●

●●●●●

●●

●●

●●

●●

●●

●●

●●

●● ● ●

●●

●●

●●●

●● ●

●●●

●●

● ●●

●●

●● ●

●●●

●●●

●●●

●●●

●●

●●●●

●●● ●●

●●

●● ●

●●●●

●● ●

●●

●●

●●

●●

●●

●●●

●●●

● ●● ●

●●

●●●●

●●

●●

●● ● ● ●●●

●●

●●

●●●●

●●

●●

●● ●

●●●●●●

●●●

●●

●●

●● ● ●

● ●●

●●●

● ●●

●●●●●

●●

●●

●●●

●●●●

●●

●●

●● ●●

●●

●●

●●●●

●●

● ●●●

●●

●●

●●●

●●

●●

● ●● ●

●●●

●●

●●●●

●●

● ●

● ●

●●

●●

● ●

●● ● ●

●●●●

●●

●●

● ● ●

●●

●● ●

●● ● ●

●●

●●

●●

●●

●●

●● ●●

●●●

●●

● ●●

●● ●

●●●●●●●

●●●

●●

● ●

●●

●●

●●●●●●

●●

●●●

●●

●●

●●

●●●

●●

●●

●●

● ● ● ● ●●●●

●●●●

●● ● ●●●●●

●●

●●●

●●

● ●

●●

●●●●

●●

●●

●●

●●

●●

●●

●●●

● ●

●●

●●●●●

●● ●

●●●●●●

● ●●● ●

●●●

● ●●●●

●●● ●●●●●

●●●

●● ●●

●●

●●●●

●●● ●●●●

●● ●●●

●●●●●●

●● ●

●●●

● ●●

● ●●●●

●●●

●●●●

● ●

●●

●●

●●●

●●

●●●

●●

●● ●

●●

●●

●●●●●

●●

●● ●

●●

●●

●●

●●●

●●

●●●●

●●

●● ●

●●

●●●●●●

● ●

● ●

●●

●●

●●

●●

● ●

● ●

●●●

●●●

●●●

●●

●● ●

●●●●●

●●

●●

●● ●

●●●

●●

●●●

●●●●●●●

● ● ●●●

●●●●●●●●

●●●● ●●

●● ● ● ● ●●●

●●●●●●

●●

●●●●●

●●

●●●

●● ●

●●

●●

● ●●

●●●

● ●●●

●●

●●

●●●

● ●● ●●●

●●●

●●●●●

●●

● ●●●●●●

●●

●●

●●

●● ●●●●

●●●

● ●●

●●●

●●

●● ●

●●●

●●●

●●●●

● ●● ●

●●

●●

●●●

●●●

● ●

●●●●●

●●

●●●

●●●●

● ●●

●●

●●● ●

● ●●

●●●

●●

●●

●●●●

●●

●●●

●●

●●

●●●

●●●

●●●●

●●

● ● ●●●●

●●

●●

●●

● ● ●

● ●●

●●

●●

●●

● ●● ●●

●●●●

●●●

●●●●

●●

●●●

●●

● ● ●●

●●

●●

●● ●

●●

●●

●●

●●

●●●

●●

● ● ●●●●●●●

●●●

●●

●●

●●●

●●

●●●●●●

● ●●

●●●●

●●●●

● ●●

●●

●●

●●

●● ●●● ●

●●

●●

●●●

●●

●●●

●● ● ●

●●●

●●●●

●●

●● ● ●●●●

●●

●●

● ●

● ●●

●●

●●●

●●

●●

●●

●●●

●●

●● ●

●●●

● ● ● ●●●●●●● ●●● ● ●●●●●●●●●●●● ● ● ●●●● ●●●●●●●●● ● ●●●●●●●●●● ●●●● ●● ●●●● ●●● ●●●●● ●●●● ●●●●●●●● ● ●●●●●●●●●● ●●●●● ●●●● ●●●●● ●●●●●●● ● ●●● ●●●●●● ●●●●●●●●●●●●●●●●●● ●●

●● ●●

●●

●●

●●

●●●

●● ●

●●●

●●●●●●●

● ●

● ●

●●

●●●●●●●

●●

●●

●●●

●●●

●●

●●

●●

●●●●

●●

● ●●●●

●●

●●●

●●

●● ● ●

●●

●●●

●●

●●●

● ● ●●

●●●

●●

●●●

● ● ●● ● ●●

●●

●●

●●

●●

●● ●

●●●●●

●●

●●●

●●

●●

●●

●●

●●●●

●●● ●

●●

● ● ● ●

●●●●●●●●

●●

●●

●●

●●●●●

●● ●●

●●

●●

●●●●●●●

●●

●●●●●

●●●●●●●●

●● ●

●●●●●●

●●

● ● ●● ●

●●

●●

●●●

●● ●

●●●

●●●●●●● ●●●●●●●● ●●●●●●●●●●● ●●● ●●●●●●●●●●● ●●●●●●● ●● ●● ●●●● ●●●● ●●●●● ●●● ●●●●●●●●●● ● ● ●●●●●●●●● ●●●●●● ●●●●● ●●●●● ●●●●● ● ● ●●●● ●●●●●●● ●●●●●●● ●●● ●●● ●● ●● ●●● ● ●●●●● ●●●●●

●●●

●● ●

●● ●

● ● ●●●

●●

●●

●●

● ●

●●

●●●

●●

●●●

●●●

●● ●

● ●●●●●●

●●

●●

● ●●

●●

●●

●●●●

●●

● ●

●●●

●●

●●

● ●

●●

●●●●

●●●●●●●

●●

●●

●● ●●

●●●

●●

●●●●

● ●●●

●●●

●●

● ●●●●

●●

●●

● ●●

●●

●●

●●●

●●

●●●

●●

●●

●●

● ● ●●●

●●

●●

●●

●●

●●●

●● ●

●●

● ●

●●

●●

●●

●●●●●

●●

●●●

●●●

●●

●●●

●●

●● ● ●●

●●

●●●●●●●

●●

● ●

●●

●●

●●

●●

● ●

●●●●

●●

●●●

●●

●●

●●

● ●●●

●●

●●●●

● ● ●●

●●●●

●●

●●●

● ● ●●

●●●●●

●●

●●●

●●

●●

●●●●●

●●●

●●

●●

●● ●●

●●●

●●

●●

●● ● ●

●●

●●

● ●●

●●●●

●●

●●

●●

●●

●●●●

● ●

● ●

● ●

●●

●●

●●●

●●●●

●●● ●

●● ●

●●●

●●●

●●

●●

●●

●●●●

●●●

●●

●●

●●●

●● ●

●●

●●

●●

●●

●●

●●●

●●●●●●●

●●

●●

●●

●●●

●●●●

●● ●

●● ●

●●

●●●●●●

●●●●●●●

●●

●●

●●

●●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●●●

●●

●●

●●

●●

●●

●●●●

●●

●●●●

●●

●●

●●●●

● ●

●●●●●

●●

●●●

● ●

●●

●●

●●

●●●●

● ●● ●

●●

●●

●●●

●●

●●

● ● ●●●

●●

●●

●● ●

●●

●●●●●●

●●

●●●●

●●

●●● ●

●●

●●

●●

●●●●●●

● ● ●●

●●

●●●

●●

●● ●

●●

●●●

●●●

●●●

●●

●●

●● ●

●●

●●●●●

●●

●●

● ●

●●

●●

●●●●

●●●●

● ● ●●●●●●

●●

●●

●●●●

●●

●●

●●

●●

●●

●●

●●

●●●●

●●

●●

●●

●●●

●●●●● ●

●●

●●●●●●

●●●

●●

●●●

● ●●

●●●●

●●

●●

● ● ● ● ●●●

●●●●●●●

●● ●

●●

●●●

●●●

●●

●●●

●●

●●

●●

●●

●●

● ●

●●●●●●

● ●●

●●

●●

●●

●●●●●

●●

● ●● ●

●●

●●●

●●

●●

● ● ● ●● ●

●●●●

●● ●●

●●

●●

●●

●●

●●

●●

●●

●●

●●

●● ●

●●●●

●●

● ●

●●

● ●

●●●

●●●●

●●

●●

●●●

●●

●●●●● ●

●●

●●●●

●●

● ●●●

●●●

●● ●● ● ●

●●●●

●●●●●●

●● ●

●●

●●

●● ● ● ●●●

●●

●●

●●

●●

●●

●●

●●

●●●●

● ●●●

● ●●

●● ●● ●

● ●●●●

●●

●●●

●●

●●

●●●●●●

●●● ● ● ●

● ●●●

●●●●●●●

●● ●

●●

●●

●●●●●

●● ● ●●

●●

●●

●●

●●

●●

●●●●● ●

●●

●●

●●●

●●●● ●●●

●●

●●●●●

●●

●●

●●●

●●

●●

●●●●●

●●●●

●●

●●●

●●

●●

●●●●

●● ●

●●●

●●●●

●●

●●●

●●

●●

●●●

●●

●●

●●

●●

●●●

●●

●●

●●●●●●

●●

●●

●●●

●●

●●

●●●

●●●

● ●● ●

●●●●

●●

● ●●

●●

●●

●●●●●

● ●

●●●

●●

●●

●●●●●●

●●

●●

●●

●●

●●●

●●

● ●

●●●

●●●●

●●●●●●●

●● ●

●●●

●●●

● ● ●● ●●●

●●

●●●●●●

●● ● ●

●●●

● ●●●●●

● ●●

●●

●●●

●●●●

●●●

●●

●●●●

●●●

●● ●

●●● ● ●

●●●

●●

●●●

● ●●

●● ● ●

●●

●●●●●●●

● ●

●●

●●●

● ●

●●

●●

●●●

●● ●

●●

●●

●●●●

●●●

●●●●

●●

●●

●●●●●●

●● ●

● ●●●

●●●●

● ●

●●● ●

●●●●●

●●●

●●

●●

● ●

●●

●●●●

●●

●●

● ●

●●

●●●●●

●●

●●

●● ●

●●●

●● ●●●●

●●

● ●

●●

● ●●● ●

●●●●

●●

●●

●●

●●

●●●

●●

●●

●●

●● ●

●●●●

●●●●●

●●●●●●

●●

● ●

●●

●●

●●●●●

●●

●●

●●

●●●

●●

●● ●

●●

●●

●●

●●●

●●

●●

●●

●●

●●

●●

●●●●

●●●

●●

●●

●●●

●●●●●

● ●

●●

●●●

●●

●●●

● ● ● ●●

● ●●

●●●●

●●

● ● ●● ● ●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●●●

●●●

●●

●●

●●●

●●

●●

●●

●●

●●●●●●

●● ●

●●

●●●

●●

●●●●

●●

●●●

●● ●●●●

● ●● ●

●●●

●●

●●

●● ●

●●●

●●●●

●●

● ●

● ●

● ●●●

●●

●●●●

●●●

●●

●●

●●

●●●●●

●●●●●●

●●

●●

●●●●●●

●● ●●

●●●

●●●

●●

●●

●●

●●●●

●●●

●●

●●

●●

● ●● ●

●●●●●

●● ● ●

●●

●●

●●

●●

●●

● ●

●●

●●●●

●●

●● ● ●

●●●●●

●●

●● ●

●●

●●●●

● ●

●●

●●●

●●

● ●●

●● ●

●●●●

●●

●●●

●●

●●

●●

●●●●●

●●

●●

● ●●

●●●●●●●

●●

●● ●● ● ●

●●

●●

●●●

●●●●

●●

●●

● ●

●●●

●●

●●●

● ●●

● ●●

●●

●●●

●●●●

● ● ●

● ●

● ●

●●

●●● ●

●●

●●

●●

●●

●●

● ●●

●●

●●

●●

●●

●●●●

●●●●

●●●

● ●●

●●●●●

●●

●●●●●

●●

●●

●● ●●

●●

●●●

●●

●●

●●

●● ●●●●

●●●

●●●●

● ●

● ●●

●●●

●●

●●

●●

●●

●● ●●

●●

● ●●

●●●

●●

●●

●●●●

●●●

●●

●●

●●

● ●●

●●

●●●

●●●●

●●●●

●●

●●

● ●

●●

●●

●●

●●

●●

●●

●●

●●●

●●

●●●

● ●

●●

● ●●

●●

●●●

●●●●●

●●

●● ●

●●

●●●●●●

●● ●

●●

●●

●●●

●●●●

●●

● ●●

●●

●●●●

●●

●●●

●●

● ●●●

●●

●●●

●●

●●

●● ●

●●

●●

●●

●● ●●

●●

●●

● ●

●● ●●●●

●●

●●

●●●

●●

●●

●●

●●

●●

●●●

●●● ●

●● ●●

●●●●

●●

●●●

●●

●●

●●●●●●

●●

●●●

●●

●●

●●

●●●●●

● ● ● ●●●●●●●●●

●●●

●●

●●

●● ●

●●

●● ●

●●

●●●●●

●●●

●●

●● ●

●●

●●

●●

●●

●●

●●

●●●

● ●●● ●

●●

●●●

●●●

●●● ●

●●●

●●●

●●

●●●●

●●

● ● ●●●●●

●●

●● ●

●●● ●●● ●

●●

● ●●

●●

●●● ●

●●

●●●

●●

●●●●●

●●

●● ●

●●

●●

●●

●●●●

●● ●

●●

●●●●

●●●●●

●●

●●

●● ●

●●

● ●

●●

●●●

●●

● ●●

●●

●●●●●

●●

●●●

●● ●●

●●

●●

●●

●●●

●●

●●

● ●●

●●●

●●

●●●

●●

●●

●●●

●●●

●●

●●

●●●●●

●●

●●●

●●●●

● ●●

● ● ●●●

●● ●

●●

●●

●●

●● ● ●

●●●

●●●●

● ●

●●● ●

●●

●●

●●

●●

●●

●●●●●●●●

●●

●●

●● ●

●● ● ●

●●●

●●

●● ● ●

●●●●

●●●

●●●●

●●

●● ●

●●●●

●●

●●

●●●

●● ●

●●●

●●

●●

●●

●●●

●● ●

●●●

●●●

●●●●

●●●●

●●●

●●●

●●

●●

●●●

● ● ● ●●●

●●

●●●●●●● ●

●● ● ● ●

●●●

●●●

●●

●●●

●●

●● ●

●●

●●●●●

●●

● ●● ● ●

●●●●

●●●●

●●●

●●● ● ●

● ●

● ●●

●● ●●

●●

●●

●●

●●

● ● ●●

●●

●●●

●●●●●

●●

●●

● ●●

●●

●●

●●

●● ●●

●●●

●●

●●

●●●●●

●●

● ●● ●

●●

●●●●●●

●●

●● ● ●

●●●

●●

●●●●

● ●

●●●●

●●

●●●●

●●

●● ● ●

●●●●

●●

●●●●

●●●

●●

●● ●

● ●●●

●●●

●●

●●●

●●●●

●●

● ●

●●

●●●

●●●●

●●

●●

●●●

●●

●●●●

●●●

●● ●

● ● ●●●

● ● ●

●●

●●

● ●

●●

●●●●

●●●

●●

● ●● ●

●●

●●●

●●●●

●●

●●

●● ●

●●

●●●●

● ●

●●

● ●●●

●●

●●

●●●●

●●

● ● ●

●●

●●

●●

●●

●●

●●

●●

●● ● ●

● ●●

●●

●●

●●

●●

●● ●

●●●●●● ●

●●●

●● ●

●●

●●

●●●

●● ●

● ●●

●●

●●●●

●●

●●

●●

● ●●

●●

●●

●●●●

●●

●● ●

●●

●●

●●

●●●●

●●

● ● ●

●●●

●●●●

●●

●●

●●

●●

●● ●● ●

●●●

●●

●●

●●

●●●

●●●●

●● ●

●●

●●●●

●●

●●● ●

●●●●

●●

●●●

●●●● ●

●●●●●

●●

●●

● ●●●●●●

●●

●●●

●●

●●

●●

● ●●●●●

●●

●●

●●

●●

● ● ●●

●●●

● ●●●

●●

● ● ● ●

● ●

●●●

●●

●●

●●

● ●

● ●●●●

●●

●●

●● ●

●●

●●●

●●

●●

●● ●

●●

●●●●●

●●

● ● ● ●●

●● ●●●

● ●

●●

●●●

●● ●

●●

●●●●●●

●●

●●

●●

●●

●●

●●●●

●●

●● ●

● ●●

●●●

●●

●●●●

● ●

●●

●●●

●●●●

●●●

●●

●●

●● ●

●●

●●

●●●

●●●

●●

● ●

●●

●●●

●●

●●

●●

● ●● ●

●●●

●●●●

● ● ●●

●●●

●●

●●●

●●●●

●●

● ●●

● ●● ●●

●●●●

●●

●● ● ●

● ●●●

●●

●●

●●

●●

● ● ●●●

●●

●●

●●

● ● ●

●●●

●●

●●●●

●●

●●●

●●

●●

● ●● ●●●

●●

●●●●

●●● ●

●●

●●●●●

●●●

●●●●●

●●●

● ●●● ●

●●

●●●

● ●●●●●●

● ●●●●●●●●

●●●●●

●●

●●

●●●

●●●●●●

●●

●●●●●●

●●

●●●● ●

●●●

●●

●●

●●

●● ●

●●

●●●

●●

●●●

●●

●●●

●●●

● ●●●●●

●●●●

●● ●

●●●

●●

●●

●●●● ● ● ● ● ● ● ●

●●●

●●●

● ●

●●

●●

●●

●●

●●

●●

●●●

●●

●●

●●

●●●

●●●

●●

●●●

● ● ● ●

●●

●●●●●

●●●●

● ● ● ●●

●●●

●●●

●●

●●●

●●

● ●●

● ●●

●●

●●●●

●●

●● ●

●●●

●●

● ●●●●●

●●

●● ●

●●

●●

−2 −1 0 1 2 3

−1

01

23

45

model.output

mon

itor.d

ata

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●● ● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●

●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ● ● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●

●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●

●●●●●●●●●●●●●●●●●●

●●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

●●●●●●●●●●●●●●●●

●●

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●

0 5000 10000 15000 20000

050

0010

000

1500

020

000

fModel

fMon

itor

Note: high correlation does not imply tail dependence!

30

Page 31: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Describing tail dependence

Limit result: extremes occur according to a Poisson process.

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

r/t[,1]

r/t[,2]

r

w

Intensity measure factors into ‘radial’ component (r) and ‘an-gular’ component (w).

Dependence is described by a probability measure on w.

31

Page 32: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Pineapple Express project

32

Page 33: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Pineapple Express project

What this section is not about

33

Page 34: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Pineapple Express project

PE storms: caused by atmospheric rivers hitting the westcoast in winter

• Often bring heavy rain and warm temperatures• Great impact on water resources of western US

Question of Interest: How well are Regional Climate Modelsable to represent extreme precipitation caused by this phe-nomenon?

34

Page 35: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Regional Climate Models

Use input from a low-resolution global model and knownphysics of the Earth system to produce simulated weatherover long periods of time at finer scales (∼50km).

NARCCAP domain

Aim: study regional impacts of climate change scenarios

For evaluation, RCMs forced by reanalysis for 1979-2004.

35

Page 36: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

RCM output vs. observations

Idea: compare daily precipitation amounts from RCM toobserved data

• Identify region and quantitythat capture PE events

• NDJF days 1981-1999

• Right: January 1, 1997 (bigPE event)

• Method: bivariate extremevalue analysis

-124 -121 -118

3540

4550

WRF

-124 -121 -118

3540

4550

Obs

0

50

100

150

200

250

36

Page 37: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Bivariate extremes analysis

Transform each marginal to unit Frechet and examinedependence

0 500 1000 1500 2000 2500

010000

20000

30000

Original Scale

WRFG Output

Obs

erve

d P

reci

p

Pineapple Express DaysNon-PE Days

0 500 1000 1500 2000 2500

0500

1000

1500

2000

2500

Frechet Scale

WRFG Output

Obs

erve

d P

reci

p

0 50 100 150 200

050

100

150

200

Frechet scale

WRF Output

Obs

erve

d P

reci

pita

tion

Histogram of w for r > r0

w

Density

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.5

1.0

1.5

Logistic modelDirichlet model

Find strong tail dependence - the WRF model representsthe largest events quite well

Not all ‘extreme’ events are PE - aim to link to processes

37

Page 38: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Pineapple Express precipitation index

Developed a PE index from daily sea-level pressure fields

-150 -140 -130 -120 -110

3035

4045

5055

60

Mean Anomaly on Extreme Precip PE days

-150 -140 -130 -120 -11030

3540

4550

5560

Mean Anomaly on Extreme Precip non-PE days

-1500

-1000

-500

0

A first step toward linking regional extreme precipitation tolarge-scale processes

38

Page 39: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Future PE events from Regional Climate Models

Use fitted dependence model to study future extreme events

0 500 1000 1500 2000 2500 3000

010000

20000

30000

40000

xDat

yDat

2040 2045 2050 2055 2060 2065 2070

15000

25000

35000

45000

Year

Sim

ulat

ed P

reci

pita

tion

Foot

prin

t

-1

0

1

2

3

Findings: increase in both frequency and intensity of PEstorms as produced by the WRF regional model forced byCCSM global model

39

Page 40: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Summary

• ‘Usual’ distributions don’t always capture tails correctly.

• Aim of EVT: study the tail of a distribution.

• An extreme value analysis utilizes only data considered‘extreme’.

• Goal of the analysis is often extrapolation.

• Foundation provided by results from probability theory.

• Communication of uncertainty is critical.

• Dependence not described with correlations/covariances.

• Application areas: climate, finance, engineering, ...

40

Page 41: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why graduate school?

• More and better job opportunities

– Often more flexible and interesting

– Not just a ‘number cruncher’

• Almost every day presents a new challenge

• Flexible (although busy) schedule

• You still get to be a student!

2009: Colorado State 23, Colorado 17 (in Boulder)

41

Page 42: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Why Statistics?

• NY Times article: “the sexy job in the next ten years”

• Data is abundant, but people who can analyze it properlyare not

• Computing power allows for new solutions to difficultproblems

– Need to quantify uncertainty in computer simulations

– SAMSI 2011-2012 UQ program

• Wide variety of applications - can ‘play in everyone else’sbackyard’

42

Page 43: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Preparing for grad school while at Concordia

As a Concordia mathematics major...

• Talk to your professors - they’re always willing to help!

• Explore research opportunities

– Research experiences for undergraduates (REU)

– Undergraduate research with Concordia professors

– SAMSI Undergraduate Workshop: February 24-25

• Take real analysis & other proof-based courses

• Get experience teaching/tutoring

When looking at schools...

• Ask a lot of questions!

• Money is important too - funding, fees, insurance, etc.

• If possible, make a visit

43

Page 44: Flooding to Financial Disaster: An Introduction to Extreme ...grantbweller.com/Presentations/wellerConcordia.pdf · #2 in season rush yds and tds 2007 vs. Carleton Mathematics & Economics

Reference and contact

Weller G., Cooley D., Sain S. An investigation of thepineapple express phenomenon via bivariate extreme valuetheory. Submitted.

Website: www.stat.colostate.edu/∼weller

Email: [email protected]

44