ANALISIS DISKRIMINAN LINEAR ROBUST METODE WMOM (WINSORIZED MODIFIED ONE-STEP M-ESTIMATORS)

The development of capital market causing more people interested to investing in stocks. Investors certainly want to invest in companies that are stable and growing. Bursa Efek Indonesia (BEI) launched the LQ45 stock index which is considered the best stock collection and non-LQ45 companies can also be another option for investors by considering the company's valuation. Financial ratios in the company's financial statements are needed to find indicators that have a major influence on the company's valuation. The method used in classifying stock groups is linear discriminant analysis. This study aims to classify data on company stock financial ratios based on LQ45 and non-LQ45 stock groups, and to determine the ratios that have a major influence on the classification. Linear discriminant analysis must fulfill the assumptions of normality and homogeneity of the variance-covariance matrix. Because the data did not meet the assumptions, a robust linear discriminant analysis using the WMOM method was used. The result is a robust linear discriminant function with a classification accuracy of 75%. Furthermore, based on the Press's Q Statistic test, the robust linear discriminant function formed is accurate and stable