PENGARUH RETURN ON EQUITY DAN DEBT TO TOTAL ASSETS RATIO TERHADAP EARNING PER SHARE (STUDI EMPIRIS PADA PERUSAHAAN MANUFAKTUR SUB SEKTOR INDUSTRI DASAR DAN KIMIA YANG TERDAFTAR DI BURSA EFEK INDONESIA (TAHUN 2016 SAMPAI TAHUN 2018)
The results of the classical assumption test based on the normality test using the Kolmogorov Smirnov test show that the significance value is 0.057, which indicates that the value is greater than 0.05, so it can be concluded that the residual data is normally distributed. The calculated dU value in this study was 1.550 and the dL value was 1.206. then the D-W count is greater than 1,550 and less than the value (4-dU) of 2,450. So that the results of the study are 1,550 <2,363 <2,450, which states that the autocorrelation coefficient is zero, meaning there is no autocorrelation.The multicollonearity test results resulted in the tolerance calculation for the variables X1 (ROE) and X2 (DAR) of 0.874, which indicates that the tolerance value for each variable is more than 0.10, which means there is no correlation between the independent variables. While assessed from the results of the valance inflation factor (VIF) on the multicolonierity test, it shows that each independent variable X1 (ROE) and X2 (DAR) of 1,144 has a VIF value below 10. So from this explanation it can be concluded that there is no multicolonierity between the independent variables in regression model. And from the scatterplots graph on financial performance on the results of the Heteroscedasticity test, it can be seen that the dots are randomly spread out over and below 0 on the Y axis, so it can be concluded that there is no heteroscedasticity.From the results above, it is explained that the variables Return On Equity (ROE) and Debt to Asset Ratio (DAR) have positive and negative effects on Earnings Per Share (EPS).