- Advanced Excel Functions Tutorial
- Advanced Excel Functions - Home
- Compatibility Functions
- Advanced Excel Functions - Cube
- Database Functions
- Date & Time Functions
- Engineering Functions
- Financial Functions
- Information Functions
- Advanced Excel Functions - Logical
- Lookup & Reference Functions
- Math & Trignometric Functions
- Statistical Functions
- Useful Resources
- Quick Guide
- Useful Resources
- Discussion
Compatibility - CHITEST Function
The CHITEST function replaces the CHISQ.TEST function in Excel 2010.
Description
The function returns the test for independence. CHITEST returns the value from the chi-squared (X2) distribution for the statistic and the appropriate degrees of freedom. You can use X2 tests to determine whether hypothesized results are verified by an experiment.
Syntax
HITEST (actual_range,expected_range)
Arguments
Argument | Description | Required/ Optional |
---|---|---|
Actual_range | The range of data that contains observations to test against expected values. | Required |
Expected_range | The range of data that contains the ratio of the product of row totals and column totals to the grand total. | Required |
Notes
The $X^2$ test first calculates a $X^2$ statistic using the formula −
$$X^2 = \sum_{i=1}^{r}\sum_{j=1}^{c}\frac{(A_{ij} - E_{ij})^2}{E_{ij}}$$
A low value of $X^2$ is an indicator of independence. As can be seen from the formula, $X^2$ is always positive or 0, and is 0 only if $A_{ij} = E_{ij}$ for every $i,j$.
CHITEST returns the probability that a value of the X2 statistic at least as high as the value calculated by the above formula could have happened by chance under the assumption of independence. In computing this probability, CHITEST uses the X2 distribution with an appropriate number of degrees of freedom, df. If r > 1 and c > 1, then df = (r - 1)(c - 1). If r = 1 and c > 1, then df = c - 1 or if r > 1 and c = 1, then df = r - 1. (r = c = 1) is not allowed and #N/A is returned.
If actual_range and expected_range have a different number of data points, CHITEST returns the #N/A error value.
Use of CHITEST is most appropriate when the values of $E_{ij}$ are not too small. Some statisticians suggest that each $E_{ij}$ should be greater than or equal to 5.
Where −
$A_{ij}$ = actual frequency in the i-th row, j-th column
$E_{ij}$ = expected frequency in the i-th row, j-th column
$r$ = number of rows
$c$ = number of rows