This macro performs a Goodness-of-Fit test between observed frequencies and
a hypothesized discrete probability distribution, using both the chi-square and
likelihood-ratio (G) statistics. Exact p-values can be requested for small
sample sizes when no parameters are estimated. Otherwise, the chi-square
approximation is used.
Download the Macro
Be sure that Minitab knows where to find your downloaded macro. Choose
.
Under
Macro location
browse to the location where you save macro files.
Important
If you use an older web browser, when you click the
Download button, the file may open in Quicktime,
which shares the .mac file extension with Minitab macros. To save the macro,
right-click the
Download button and choose
Save target as.
Required Inputs
- Column containing the
observed frequencies (integer counts).
- Column containing either
the expected numbers or the hypothesized probabilities. If expected numbers are
used, the sum of values in this column must equal the sample size. If
probabilities are used, the sum must equal 1.0.
Optional Inputs
- ESTIMATE K
- Use to specify the number (K) of parameters estimated from the data
so that the degrees of freedom of the test statistic can be decreased
accordingly. The default value for K is zero.
- EXACT
- Use to calculate an exact P-value for the chi-square statistic based
on the multinomial distribution. The exact test should be invoked only when
there are relatively few categories (2 or 3) and/or the sample sizes are so
small that the chi-square approximation is unreliable. Because this routine
invokes built-in Minitab functions many times, it is highly inefficient and
will require a large amount of time to execute if sample sizes or the number of
categories are too large. In the special case of two categories, the chi-square
statistic is calculated both without and with the Yates correction for
continuity.
Running the Macro
Suppose the observed frequencies are in C1, the expected numbers are in
C2. To run the macro, choose
and type:
%GOF C1 C2
Click
Run.
