Open ipbyrne opened 6 years ago
1-way classification: Goodness-of-fit test X^2 Formula X^2 = ((Observed Value - Expected Value)^2/Expected Value) + ((Observed Value - Expected Value)^2/Expected Value);
// Calculated df (degrees of freedom) df = k-1, k = number of categories (2 by default - two swing outcomes)
// Find p-value X^2(df)
Link to table based of df value: http://ib.bioninja.com.au/_Media/chi-table_med.jpeg
Check for X^2 value that produces a p value of 0.05 or less and 0.01 or less.
Want to add array storing the Chi-Square table to get more exact p-values
Also need to add chi-squared info/formula to READ.ME
Look to run ANOVA test as well to cross verify.
Currently, perform the swing data verification inside an excel spreadsheet. Need to implement the test into the indicator itself.
Chi-Squared Test:
LINK ONE: https://web.stanford.edu/class/psych252/cheatsheets/chisquare.html
LINK TWO: http://onlinestatbook.com/2/chi_square/one-way.html
LINK THREE: http://onlinestatbook.com/2/chi_square/distribution.html
Alternatives: One-Way ANOVA KS test: https://en.wikipedia.org/wiki/Kolmogorov%E2%80%93Smirnov_test Shapiro-Wilk test: https://en.wikipedia.org/wiki/Shapiro%E2%80%93Wilk_test Anderson-Darling test: https://en.wikipedia.org/wiki/Anderson%E2%80%93Darling_test