MATLAB Using gamfit Function & LR Hypothesis Test in MATLAB

AI Thread Summary
The discussion centers on fitting data to a gamma distribution using MATLAB's gamfit function, which provides maximum likelihood estimates (MLEs) and 95% confidence intervals for the parameters. The user seeks to compute the standard errors of the coefficients instead of relying on the confidence intervals. A formula is mentioned for calculating standard error based on the confidence interval, specifically that it can be derived from the difference between the upper and lower limits divided by 3.92. Additionally, the user expresses interest in performing a likelihood ratio hypothesis test but is unsure how to obtain the necessary NullLLF and BaseLLF values for input into the lratiotest function.
joinlia
Messages
1
Reaction score
0
Hello,
I fit my data to gamma distribution using the gamfit function of MATLAB.
The gamfit function returns MLEs and 95% percent confidence intervals as follows:
• a = 2; b = 4;data = gamrnd(a,b,100,1);
• [p,ci] = gamfit(data)
• p =
• 2.1990 3.7426
• ci =
• 1.6840 2.8298
• 2.7141 4.6554
I am interested in finding the standard error of the coefficients rather than the confidence interval.
I want to know how to compute the standard errors based on the confidence interval??

Moreover, likelihood ratio hypothesis test were used to test hypotheses such as p(1)=0.5, but I don’t know how to obtaine and input the NullLLF and BaseLLF values

Syntax likelihood ratio hypothesis test
[H,pValue,Ratio,CriticalValue]=lratiotest(BaseLLF,NullLLF,DoF,Alpha)
 
Physics news on Phys.org
From a quick google search, you can find that the standard error = (upper limit – lower limit) / 3.92.
 
Last edited:

Similar threads

Back
Top