Difference between revisions of "Manuals/calci/LEVENESTEST"

From ZCubes Wiki
Jump to navigation Jump to search
(Created page with "creation of levenes test.")
 
 
(33 intermediate revisions by 4 users not shown)
Line 1: Line 1:
−
creation of levenes test.
+
<div style="font-size:30px">'''LEVENESTEST (DataRange,ConfidenceLevel,NewTableFlag)'''</div><br/>
 +
*<math>DataRange</math> is the set of values for the test.
 +
*<math>ConfidenceLevel</math> is the value from 0 to 1.
 +
*<math>NewTableFlag</math> is either TRUE or FALSE. TRUE for getting results in a new cube. FALSE will display results in the same cube.
 +
 
 +
==Description==
 +
*This function used to test the Homogeneity of variances.
 +
*Levene's test is used to test the Samples have equal variances.
 +
*Equal variances across samples is called homogeneity of variance or homoscedasticity.
 +
*To do the Levenes test we need the following assumptions:
 +
  1.The Samples from the populations are independent of one another.
 +
  2. The population under consideration are Normally Distributed.
 +
*For three or more variables the following statistical tests for homogeneity of variances are commonly used:
 +
    1.Levene's Test.
 +
    2.Bartlett Test.
 +
*Levene's test is an alternative to the Bartlett test.
 +
*If the data surely is of normally distributed or nearly to normally distributed then  we can use the Bartlett test.
 +
*The Levene's test is defined as
 +
<math>H_0 = \sigma_1^2 = \sigma_2^2=...... = \sigma_t^2</math>.
 +
<math>H_1</math>=Not all of the variances are equal.
 +
*Normally there are three versions of the Levenes test.
 +
*There are
 +
*1.Use of Mean.
 +
*2.Use of Median.
 +
*3.Use of  10% of Trimmed Mean.
 +
*The Levene test statistic is:
 +
<math>W=\frac{(N-k)\sum_{i=1}^k N_i(Z_i-Z)^2}{(k-1)\sum_{i=1}^k \sum_{i=1}^k \sum_{j=1}^{N_i} (Z_{ij}-Z_i)^2}</math>.
 +
**where <math>W</math> is the result of the test.
 +
**<math>k</math> is the number of different groups to which the sampled cases belong.
 +
**<math>N</math> is the total number of cases in all groups.
 +
**<math>N_i</math>  is the number of cases in the <math>i^{th}</math> group.
 +
**<math>Y_{ij} is the value of the measured variable for the <math>j^{th}</math> case from the <math>i^{th}</math> group.
 +
*Zij is satisfying the one of the following conditions:
 +
*1.<math>z_{ij}=|y_{ij}-\bar{y_i}|</math>,Where <math>\bar{y_i}</math> is the Mean of the <math>i^{th}</math> subgroup.
 +
*2.<math>z_{ij}=|y_{ij}-\bar{y_i}|</math>,Where <math>\bar{y_i}</math> is the Median of the <math>i^{th}</math> subgroup
 +
*3.<math>z_{ij}=|y_{ij}-\bar{y_i}|</math>,Where <math>\bar{y_i}</math> is the 10%Trimmed Mean of the <math>i^{th}</math> subgroup.
 +
*Levene's Testing Procedure:
 +
*1. checking the assumptions.
 +
*2.State the Null(H0) and alternative(H1) hypothesis.
 +
*3.Decide on the Significance level (α).
 +
*4.Finding the Critical value and Rejection Region.Here <math>df_1=t-1</math>,<math>df_2=N-t</math>.
 +
*5.Compute the Levenes statistic using the formula.
 +
*6.Then decision of  the value of the test statistic,W is falls in the rejection region or if p-value ≤ α, then reject <math>H_0</math>.Otherwise, fail to reject <math>H_0</math>. For the computation p-value we have to use the value of <math>df_1</math> and <math>df_2</math>.
 +
*7. Finally we have to conclude that the rejection of <math>H_0</math> or fail to rejection <math>H_0</math> according to the test statistic at the significance level.
 +
 
 +
==Example==
 +
{| class="wikitable"
 +
|-
 +
| X1 || X2
 +
|-
 +
| 3067 || 3200
 +
|-
 +
| 2730 || 2777
 +
|-
 +
| 2840 || 2623
 +
|-
 +
| 2913 || 3044
 +
|-
 +
| 2789 || 2834
 +
|}
 +
*=LEVENESTEST(B1:C5,.05,0)
 +
{| class="wikitable"
 +
|+LEVENES TEST
 +
|-
 +
!  !! DATA-0 !! DATA-1
 +
|-
 +
| Median || 2840 || 2834
 +
|-
 +
| Mean || 2867.8 || 2895.6
 +
|-
 +
| Variance || 16923.7 || 51713.3
 +
|-
 +
| Count || 5 || 5
 +
|-
 +
| df || 4 || 4
 +
|}
 +
 
 +
{| class="wikitable"
 +
|+SUMMARY OUTPUT
 +
|-
 +
! LEVENESTEST !! STATISTICS
 +
|-
 +
| W || 1.0439235110342522
 +
|-
 +
| F-Test || 0.38245649772919
 +
|-
 +
| a || 0.05
 +
|-
 +
| F || 0.32726010523405
 +
|-
 +
| p 1 & 2 Tail || 0.1524069466470822 || 0.3048138932941644
 +
|}
 +
 
 +
==Related Videos==
 +
 
 +
{{#ev:youtube|81Yi0cTuwzw|280|center|Levene's Test}}
 +
 
 +
==See Also==
 +
*[[Manuals/calci/SIGNTEST| SIGNTEST]]
 +
*[[Manuals/calci/FRIEDMANTEST| FRIEDMANTEST]]
 +
 
 +
 
 +
==References==
 +
 
 +
*[http://en.wikipedia.org/wiki/Levene%27s_test Levene's test documentation on Wikipedia]
 +
*[http://www.qimacros.com/hypothesis-testing/levenes-test/ Levene's test for variance in Excel]

Latest revision as of 15:57, 14 June 2018

LEVENESTEST (DataRange,ConfidenceLevel,NewTableFlag)


  • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle DataRange} is the set of values for the test.
  • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle ConfidenceLevel} is the value from 0 to 1.
  • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle NewTableFlag} is either TRUE or FALSE. TRUE for getting results in a new cube. FALSE will display results in the same cube.

Description

  • This function used to test the Homogeneity of variances.
  • Levene's test is used to test the Samples have equal variances.
  • Equal variances across samples is called homogeneity of variance or homoscedasticity.
  • To do the Levenes test we need the following assumptions:
  1.The Samples from the populations are independent of one another. 
  2. The population under consideration are Normally Distributed.
  • For three or more variables the following statistical tests for homogeneity of variances are commonly used:
   1.Levene's Test.
   2.Bartlett Test.
  • Levene's test is an alternative to the Bartlett test.
  • If the data surely is of normally distributed or nearly to normally distributed then we can use the Bartlett test.
  • The Levene's test is defined as
Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_0 = \sigma_1^2 = \sigma_2^2=...... = \sigma_t^2}
.
Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_1}
=Not all of the variances are equal.
  • Normally there are three versions of the Levenes test.
  • There are
  • 1.Use of Mean.
  • 2.Use of Median.
  • 3.Use of 10% of Trimmed Mean.
  • The Levene test statistic is:

Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle W=\frac{(N-k)\sum_{i=1}^k N_i(Z_i-Z)^2}{(k-1)\sum_{i=1}^k \sum_{i=1}^k \sum_{j=1}^{N_i} (Z_{ij}-Z_i)^2}} .

    • where Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle W} is the result of the test.
    • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle k} is the number of different groups to which the sampled cases belong.
    • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle N} is the total number of cases in all groups.
    • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle N_i} is the number of cases in the Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle i^{th}} group.
    • Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle Y_{ij} is the value of the measured variable for the <math>j^{th}} case from the Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle i^{th}} group.
  • Zij is satisfying the one of the following conditions:
  • 1.Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle z_{ij}=|y_{ij}-\bar{y_i}|} ,Where Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle \bar{y_i}} is the Mean of the Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle i^{th}} subgroup.
  • 2.Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle z_{ij}=|y_{ij}-\bar{y_i}|} ,Where Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle \bar{y_i}} is the Median of the Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle i^{th}} subgroup
  • 3.Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle z_{ij}=|y_{ij}-\bar{y_i}|} ,Where Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle \bar{y_i}} is the 10%Trimmed Mean of the Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle i^{th}} subgroup.
  • Levene's Testing Procedure:
  • 1. checking the assumptions.
  • 2.State the Null(H0) and alternative(H1) hypothesis.
  • 3.Decide on the Significance level (α).
  • 4.Finding the Critical value and Rejection Region.Here Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle df_1=t-1} ,Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle df_2=N-t} .
  • 5.Compute the Levenes statistic using the formula.
  • 6.Then decision of the value of the test statistic,W is falls in the rejection region or if p-value ≤ α, then reject Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_0} .Otherwise, fail to reject Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_0} . For the computation p-value we have to use the value of Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle df_1} and Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle df_2} .
  • 7. Finally we have to conclude that the rejection of Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_0} or fail to rejection Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://wikimedia.org/api/rest_v1/":): {\displaystyle H_0} according to the test statistic at the significance level.

Example

X1 X2
3067 3200
2730 2777
2840 2623
2913 3044
2789 2834
  • =LEVENESTEST(B1:C5,.05,0)
LEVENES TEST
DATA-0 DATA-1
Median 2840 2834
Mean 2867.8 2895.6
Variance 16923.7 51713.3
Count 5 5
df 4 4
SUMMARY OUTPUT
LEVENESTEST STATISTICS
W 1.0439235110342522
F-Test 0.38245649772919
a 0.05
F 0.32726010523405
p 1 & 2 Tail 0.1524069466470822 0.3048138932941644

Related Videos

Levene's Test

See Also


References

Retrieved from "http://wiki.zcubes.com/index.php?title=Manuals/calci/LEVENESTEST&oldid=210292"