Tests for heteroscedasticity, Advanced Statistics

Lagrange Multiplier (LM) test

The Null Hypothesis - H0: There is no heteroscedasticity i.e. β1 = 0

The Alternative Hypothesis - H1:  There is heteroscedasticity i.e. β1 0

Reject H0 if nR2 > 2094_Tests for Heteroscedasticity.png

Regression Analysis: sqresi versus sqfits

The regression equation is

sqresi = 0.00517 + 0.0196 sqfits

Predictor    Coef          SE Coef       T       P        VIF

Constant   0.005173  0.001130  4.58  0.000

sqfits         0.019650  0.008395  2.34  0.019  1.000

 

S = 0.0112996   R-Sq = 0.4%   R-Sq(adj) = 0.3%

Analysis of Variance

Source             DF         SS                   MS           F      P

Regression       1        0.0006996  0.0006996  5.48  0.019

Residual Error  1500  0.1915214  0.0001277

  Lack of Fit      646    0.0819554  0.0001269  0.99  0.559

  Pure Error     854    0.1095659   0.0001283

Total           1501  0.1922209

MTB > let k1 = 1502*0.04

MTB > print k1

Data Display

K1    60.0800

Inverse Cumulative Distribution Function

Chi-Square with 1 DF

P( X <= x )        x

       0.95  3.84146

Since nR2 = (1502*0.04) 60.0800 > 3.84146 = 2094_Tests for Heteroscedasticity.png, there is sufficient evidence to reject H0 which suggest that there is heteroscedasticity from the Lagrange Multiplier (LM) test at 5% significance level which means that one or more slopes are not zero.

Posted Date: 3/5/2013 5:59:53 AM | Location : United States







Related Discussions:- Tests for heteroscedasticity, Assignment Help, Ask Question on Tests for heteroscedasticity, Get Answer, Expert's Help, Tests for heteroscedasticity Discussions

Write discussion on Tests for heteroscedasticity
Your posts are moderated
Related Questions
An oil company is considering whether or not to bid for an offshore drilling contract. If they bid, the value would be $600m with a 65% chance of gaining the contract. The company

Paired samples are the two samples of the observations with the characteristic feature with each of the observation in one sample have only one matching observation in the other s

In an experiment, power is a function of 1. The number of variables being measured and the beta level 2. The effect size, internal validity and the beta level 3. The number of part


Model is the description of the supposed structure of a set of observations which can range from a fairly imprecise verbal account to, more commonly, a formalized mathematical exp

Time series : The values of a variable recorded, generally at a regular interval, over the long period of time. The observed movement and fluctuations of several such series are

Zero sumgame is a game played by the number of persons in which the winner takes all stakes given by the losers so that the algebraic sum of gains at any stage is zero. Number of

Network sampling is a sampling design in which the simple random sample or strati?ed sample of the sampling units is made and all observational units which are linked to any of th

what is pdf,mean & variance for multimodal distribution?

Relative risk is the measure of the association between the exposure to a particular factor and the risk or probability of a convinced outcome, calculated as follows     therefor