error analysis
DESCRIPTION
The presentation deals with the aspects of error analysis and the methodologies employed in data rejection and details of determining confidence intervals.TRANSCRIPT
Error Analysis and Minimisation
Abhiram MullapudiCB.EN.U4CIE11038
08-09-2014 Seminar CHY 259 1
INTRODUCTION
DATA
Design
Analysis
Study
08-09-2014 Seminar CHY 259 2
• Error
True value(accepted value)- Measured value
• Accuracy
• Precision
08-09-2014 Seminar CHY 259 3
Expressing Accuracy
• Absolute Error
Measured value – True value
• Relative Error
𝐴𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝐸𝑟𝑟𝑜𝑟
𝑇𝑟𝑢𝑒 𝑉𝑎𝑙𝑢𝑒× 100
08-09-2014 Seminar CHY 259 4
Classification of Error
08-09-2014 Seminar CHY 259 5
• Determinate or Systematic Error
• Indeterminate or Accidental Errors
Minimization of Error
• Calibration of apparatus and application of corrections
• Running a control determination
• Use of independent methods of analysis
• Running parallel determinations
08-09-2014 Seminar CHY 259 6
Significant figures• The significant figures of a number are those digits
that carry meaning contributing to its precision
08-09-2014 Seminar CHY 259 7
1. All non-zero numbers (1,2,3,4,5,6,7,8,9) are always significant.
2. All zeroes between non-zero numbers are always significant.
3. All zeroes which are simultaneously to the right of the decimal point and at the end of the number are always significant.
4. All zeroes which are to the left of a written decimal point and are in a number >= 10 are always significant.
08-09-2014 Seminar CHY 259 8
Characteristics of Replicate Measurements
• Range
𝑅𝑎𝑛𝑔𝑒 ∝1
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛• Mean
• Median
08-09-2014 Seminar CHY 259 9
Deviation
• Mean Deviation
• Standard Deviation
It measures the amount of variation or dispersion from the average
08-09-2014 Seminar CHY 259 10
Rejection of Data
• Average Deviation
1. Omit the doubtful value
2. Determine Mean and Avg.Deviation
3. Check for
𝑥 ≥ 4 × 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
08-09-2014 Seminar CHY 259 11
Standard Deviation
08-09-2014 Seminar CHY 259 12
Dixon’s Q test
• Arrange the data in order of increasing values
• Calculate Q
𝑄 =𝐺𝑎𝑝
𝑅𝑎𝑛𝑔𝑒
• If Q > Qtable, where Qtable is a reference value corresponding to the sample size and confidence level, then reject the questionable point
08-09-2014 Seminar CHY 259 13
Confidence Limits
𝐶𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝐿𝑖𝑚𝑖𝑡 = 𝑥 ±(𝑁 − 1)𝑠
𝑁
08-09-2014 Seminar CHY 259 14
Comparison of Results
• Students t test
𝑡 =𝑥 − 𝜇 √𝑁
𝑆
Compare the t value with the value in the table.
08-09-2014 Seminar CHY 259 15
• F test
𝐹 =𝑆𝑎2
𝑆𝑏2
Compare the values to the table.
08-09-2014 Seminar CHY 259 16
Thank you.
08-09-2014 Seminar CHY 259 17