Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 65052

Linear Discriminant Analysis

$
0
0
Hello,

i'm conducting a research with active and faling companies. So my dependent variable is a categorical one with faling(0) and active(1). All my independent variables are scale variables. As i'm testing the assumptions for LDA ( equal variance, normal distribution, means, ...) i have to conclude that no test is affirmative. ( No equal variances, no normal distributions, ...) Since i think about my data it's logical that one group (faling) is totally different than the other group(active). Can i solve these assumptions? or can i ignore some assumptions? I've based my choice for this method on the literature and i would like to use this method ( Since it's used a lot in literature for the same concept). The only difference is the data and different years.

Although i have a huge dataset - so the normal distribution can be refuted- the other tests are negative.

Kind regards,

Olivier De Keyzer

Viewing all articles
Browse latest Browse all 65052

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>