# What statistical test(s) should i use to test each of the following hypotheses?

The outcome variable is financial distress with 3 categories while the independent variables are financial factors and are continuous (about 12 of them).

• $$H_0$$: Financial factors have no impact on the financial distress states.
• $$H_0$$: financial factors when controlled for Macroeconomic factors have no significant impact on financial distress states.
• $$H_0$$: Determinant factors of each of the three financial distress states categories are not different.
• The first two could be tested by likelihood ratio tests of multinomial logistic regression models, the first being a comparison of the model with all financial factors to an intercept-only model and the second a comparison of the model with financial factors and macroeconomic factors to a model with just macroeconomic factors. I don't follow exactly what you mean in the third hypothesis.
– Dave
Sep 17, 2020 at 19:00

When your predictors (i.v.) are continuous and your outcome variable (financial distress) is categorical level, most likely your best statistical test is simply logistic regression.

I hope this helps, good luck with your research!

• Logistic regression is a model and does not perform any kind of hypothesis test on its own.
– Dave
Sep 17, 2020 at 17:30
• Ah apologies, in applications such as SPSS the type of hypothesis test is simply named after the model. More specific in this case (since your dependent var has 3 categories, it would be the 'multinomial logistic regression test' Sep 17, 2020 at 17:49
• Multinomial logistic regression isn't a test, either.
– Dave
Sep 17, 2020 at 18:58