Statistics Tutorial
Selected Reading
- Statistics - Discussion
- Z table
- Weak Law of Large Numbers
- Venn Diagram
- Variance
- Type I & II Error
- Trimmed Mean
- Transformations
- Ti 83 Exponential Regression
- T-Distribution Table
- Sum of Square
- Student T Test
- Stratified sampling
- Stem and Leaf Plot
- Statistics Notation
- Statistics Formulas
- Statistical Significance
- Standard normal table
- Standard Error ( SE )
- Standard Deviation
- Skewness
- Simple random sampling
- Signal to Noise Ratio
- Shannon Wiener Diversity Index
- Scatterplots
- Sampling methods
- Sample planning
- Root Mean Square
- Residual sum of squares
- Residual analysis
- Required Sample Size
- Reliability Coefficient
- Relative Standard Deviation
- Regression Intercept Confidence Interval
- Rayleigh Distribution
- Range Rule of Thumb
- Quartile Deviation
- Qualitative Data Vs Quantitative Data
- Quadratic Regression Equation
- Process Sigma
- Process Capability (Cp) & Process Performance (Pp)
- Probability Density Function
- Probability Bayes Theorem
- Probability Multiplecative Theorem
- Probability Additive Theorem
- Probability
- Power Calculator
- Pooled Variance (r)
- Poisson Distribution
- Pie Chart
- Permutation with Replacement
- Permutation
- Outlier Function
- One Proportion Z Test
- Odd and Even Permutation
- Normal Distribution
- Negative Binomial Distribution
- Multinomial Distribution
- Means Difference
- Mean Deviation
- Mcnemar Test
- Logistic Regression
- Log Gamma Distribution
- Linear regression
- Laplace Distribution
- Kurtosis
- Kolmogorov Smirnov Test
- Inverse Gamma Distribution
- Interval Estimation
- Individual Series Arithmetic Mode
- Individual Series Arithmetic Median
- Individual Series Arithmetic Mean
- Hypothesis testing
- Hypergeometric Distribution
- Histograms
- Harmonic Resonance Frequency
- Harmonic Number
- Harmonic Mean
- Gumbel Distribution
- Grand Mean
- Goodness of Fit
- Geometric Probability Distribution
- Geometric Mean
- Gamma Distribution
- Frequency Distribution
- Factorial
- F Test Table
- F distribution
- Exponential distribution
- Dot Plot
- Discrete Series Arithmetic Mode
- Discrete Series Arithmetic Median
- Discrete Series Arithmetic Mean
- Deciles Statistics
- Data Patterns
- Data collection - Case Study Method
- Data collection - Observation
- Data collection - Questionaire Designing
- Data collection
- Cumulative Poisson Distribution
- Cumulative plots
- Correlation Co-efficient
- Co-efficient of Variation
- Cumulative Frequency
- Continuous Series Arithmetic Mode
- Continuous Series Arithmetic Median
- Continuous Series Arithmetic Mean
- Continuous Uniform Distribution
- Comparing plots
- Combination with replacement
- Combination
- Cluster sampling
- Circular Permutation
- Chi Squared table
- Chi-squared Distribution
- Central limit theorem
- Boxplots
- Black-Scholes model
- Binomial Distribution
- Beta Distribution
- Best Point Estimation
- Bar Graph
- Arithmetic Range
- Arithmetic Mode
- Arithmetic Median
- Arithmetic Mean
- Analysis of Variance
- Adjusted R-Squared
- Home
Selected Reading
- Who is Who
- Computer Glossary
- HR Interview Questions
- Effective Resume Writing
- Questions and Answers
- UPSC IAS Exams Notes
Kurtosis
Statistics - Kurtosis
The degree of tailedness of a distribution is measured by kurtosis. It tells us the extent to which the distribution is more or less outper-prone (heavier or pght-tailed) than the normal distribution. Three different types of curves, courtesy of Investopedia, are shown as follows −
It is difficult to discern different types of kurtosis from the density plots (left panel) because the tails are close to zero for all distributions. But differences in the tails are easy to see in the normal quantile-quantile plots (right panel).
The normal curve is called Mesokurtic curve. If the curve of a distribution is more outper prone (or heavier-tailed) than a normal or mesokurtic curve then it is referred to as a Leptokurtic curve. If a curve is less outper prone (or pghter-tailed) than a normal curve, it is called as a platykurtic curve. Kurtosis is measured by moments and is given by the following formula −
Formula
${eta_2 = frac{mu_4}{mu_2}}$
Where −
${mu_4 = frac{sum(x- ar x)^4}{N}}$
The greater the value of eta_2 the more peaked or leptokurtic the curve. A normal curve has a value of 3, a leptokurtic has eta_2 greater than 3 and platykurtic has eta_2 less then 3.
Example
Problem Statement:
The data on daily wages of 45 workers of a factory are given. Compute eta_1 and eta_2 using moment about the mean. Comment on the results.
| Wages(Rs.) | Number of Workers |
|---|---|
| 100-200 | 1 |
| 120-200 | 2 |
| 140-200 | 6 |
| 160-200 | 20 |
| 180-200 | 11 |
| 200-200 | 3 |
| 220-200 | 2 |
Solution:
| Wages (Rs.) | Number of Workers (f) | Mid-pt m | m-${frac{170}{20}}$ d | ${fd}$ | ${fd^2}$ | ${fd^3}$ | ${fd^4}$ |
|---|---|---|---|---|---|---|---|
| 100-200 | 1 | 110 | -3 | -3 | 9 | -27 | 81 |
| 120-200 | 2 | 130 | -2 | -4 | 8 | -16 | 32 |
| 140-200 | 6 | 150 | -1 | -6 | 6 | -6 | 6 |
| 160-200 | 20 | 170 | 0 | 0 | 0 | 0 | 0 |
| 180-200 | 11 | 190 | 1 | 11 | 11 | 11 | 11 |
| 200-200 | 3 | 210 | 2 | 6 | 12 | 24 | 48 |
| 220-200 | 2 | 230 | 3 | 6 | 18 | 54 | 162 |
| ${N=45}$ | ${sum fd = 10}$ | ${sum fd^2 = 64}$ | ${sum fd^3 = 40}$ | ${sum fd^4 = 330}$ |
Since the deviations have been taken from an assumed mean, hence we first calculate moments about arbitrary origin and then moments about mean. Moments about arbitrary origin 170
${mu_1^1= frac{sum fd}{N} imes i = frac{10}{45} imes 20 = 4.44 \[7pt] mu_2^1= frac{sum fd^2}{N} imes i^2 = frac{64}{45} imes 20^2 =568.88 \[7pt] mu_3^1= frac{sum fd^2}{N} imes i^3 = frac{40}{45} imes 20^3 =7111.11 \[7pt] mu_4^1= frac{sum fd^4}{N} imes i^4 = frac{330}{45} imes 20^4 =1173333.33 }$Moments about mean
${mu_2 = mu _2 - (mu _1 )^2 = 568.88-(4.44)^2 = 549.16 \[7pt] mu_3 = mu _3 - 3(mu _1)(mu _2) + 2(mu _1)^3 \[7pt] , = 7111.11 - (4.44) (568.88)+ 2(4.44)^3 \[7pt] , = 7111.11 - 7577.48+175.05 = - 291.32 \[7pt] \[7pt] mu_4= mu _4 - 4(mu _1)(mu _3) + 6 (mu_1 )^2 (mu _2) -3(mu _1)^4 \[7pt] , = 1173333.33 - 4 (4.44)(7111.11)+6(4.44)^2 (568.88) - 3(4.44)^4 \[7pt] , = 1173333.33 - 126293.31+67288.03-1165.87 \[7pt] , = 1113162.18 }$From the value of movement about mean, we can now calculate ${eta_1}$ and ${eta_2}$:
${eta_1 = mu^2_3 = frac{(-291.32)^2}{(549.16)^3} = 0.00051 \[7pt] eta_2 = frac{mu_4}{(mu_2)^2} = frac{1113162.18}{(546.16)^2} = 3.69 }$From the above calculations, it can be concluded that ${eta_1}$, which measures skewness is almost zero, thereby indicating that the distribution is almost symmetrical. ${eta_2}$ Which measures kurtosis, has a value greater than 3, thus implying that the distribution is leptokurtic.
Advertisements