The Effect of Net Profit Margin, Debt to Equity Ratio, and Return on Equity against Company Value in Food and Beverage Manufacturing Sub-sector Companies listed on the Indonesia Stock Exchange

High and low stock prices of a company can be observed and studied with two approaches both technically and fundamentally. Technically, observations are carried out by paying attention to the charts in a coherent manner from several observations (time series) to get a picture of stock movements, but on the other hand technical analysis can also be used such as studying and observing based on the company's financial performance, demand and supply, interest rates, risk levels, inflation rates, government policies, political conditions and security of a country. However, the company's financial condition has the most dominant influence on the formation of stock prices. The company's financial report that are often used in analysis are the balance sheet and income statement. The balance sheet shows all the assets owned by a company at a point in time and the source of capital to buy these assets. Periodically public companies listed on the exchange must publish their financial statements, while the income statement provides an overview of the company's performance. Simply put, stock prices can be predicted by analyzing available financial data. Some of the methods used to measure company value are Price Earning Ratio (PER) showing how much money investors are willing to spend to pay every dollar of reported profit (Brigham and Houston, 2006: 110). This ratio is used to measure how much the ratio between the company's stock prices with the profits obtained by shareholders. The use of price earning ratio is to see how the market appreciates the company's performance which is reflected by its earnings per share. Price earning ratio shows the relationship between the ordinary stock market with earnings per share. Price earning ratio (PER) serves to measure


I. Introduction
High and low stock prices of a company can be observed and studied with two approaches both technically and fundamentally. Technically, observations are carried out by paying attention to the charts in a coherent manner from several observations (time series) to get a picture of stock movements, but on the other hand technical analysis can also be used such as studying and observing based on the company's financial performance, demand and supply, interest rates, risk levels, inflation rates, government policies, political conditions and security of a country. However, the company's financial condition has the most dominant influence on the formation of stock prices.
The company's financial report that are often used in analysis are the balance sheet and income statement. The balance sheet shows all the assets owned by a company at a point in time and the source of capital to buy these assets. Periodically public companies listed on the exchange must publish their financial statements, while the income statement provides an overview of the company's performance. Simply put, stock prices can be predicted by analyzing available financial data.
Some of the methods used to measure company value are Price Earning Ratio (PER) showing how much money investors are willing to spend to pay every dollar of reported profit (Brigham and Houston, 2006: 110). This ratio is used to measure how much the ratio between the company's stock prices with the profits obtained by shareholders. The use of price earning ratio is to see how the market appreciates the company's performance which is reflected by its earnings per share. Price earning ratio shows the relationship between the ordinary stock market with earnings per share. Price earning ratio (PER) serves to measure Lenders and investors usually choose a low Debt to Equity Ratio because their interests are better protected if there is a decline in business at the company concerned. Therefore, a company that has a high Debt to Equity Ratio may not be able to attract additional capital with loans from other parties.
The formula for calculating Debt to Equity Ratio is:

Return on Equity Theory
ROE (Return on Equity) compares net income after tax with equity that has been invested by shareholders of the company (Van Horne and Wachowicz, 2005: 225). This ratio shows the power to generate a return on investment based on the book value of shareholders, and is often used in comparing two or more companies for good investment opportunities and cost effective management. ROE is very attractive to holders and prospective shareholders, and also to management, because the ratio is an important measure or indicator of shareholders value creation, meaning that the higher the ROE ratio, the higher the company value, this is certainly an attraction for Based on what has been described previously, the conceptual framework in this study can be described as follows:

III. Research Method
This research was conducted at Food and Beverage sub-sector manufacturing companies listed on the Stock Exchange in 2014-2018 period with the address of the Stock Exchange office on Jalan Jenderal Sudirman Kav. 52 -53 Jakarta, Indonesia:

Independent Variable
a. Net Profit Margin (X1), which is the profitability ratio used to measure the percentage of net income in a company against net sales. This Net Profit Margin shows the proportion of sales remaining after deducting all related costs. The proxies used for NPM are as follows:

Return on Equity Ratio (X3)
The company value (Y) Debt to Equity Ratio (X2)

(X1)
Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www. b. Debt to Equity Ratio / DER (X2) is the ratio of the ability of food and beverage manufacturing companies to cover loans from their own capital. The proxy used for this DER is as follows: c. Return on Equity Ratio (X3) is the ratio of the ability of food and beverage manufacturing companies to generate profits using their own capital. The proxies used for ROE are as follows:

Dependent Variable
The dependent variable in this study is company value. The tool used in measuring the value of this company is Price to Book Value (Y). The price used in this case is the closing price divided by the Book Value of Shares (total equity divided by total outstanding shares). Calculation of the company value / calculation of this Price to Book Value is to find out how expensive or cheap the stock price of a company is today. Price to Book Value also consistently and precisely represents the fair company value's stock value. Because Price to Book Value is calculated based on company equity where as long as the company is able to generate profits, its value will also continue to rise.

Classical Assumption Tests
The classic assumption test is a statistical requirement that must be met in multiple linear regression analysis. Before testing the hypothesis, first testing is done whether there is a violation of classical assumptions. The results of testing a good hypothesis is testing that does not violate. These tests include: normality test, multicollinearity test, autocorrelation test, and heteroscedasticity test.

a. Normality Test
The normality test aims to test whether in the regression model, the dependent variable and the independent variable both have normal or abnormal distributions. A good regression model is to have normal or near-normal data distribution. The normality test is carried out with the SPSS 24.0 program which uses the One-Sample Kolmogorov-Smirnov Test and the Normal Propability Plot. In the Normal Propability Plot, data normality can be detected by looking at the spread of data (points) on the diagonal axis of the graph. The basis of decision making is if the data spreads around the diagonal line and follows the direction of the diagonal line, then the regression meets the assumption of normality. If the data spreads far from the diagonal line and or does not follow the direction of the diagonal line, the regression model does not meet the assumption of normality.
In the One-Sample Kolmogorov-Smirnov Test, the criteria are: 1) If Asymp. Sig. (2-tailed) <0.05, the regression model does not produce residual values with normal distribution or reject H0.

b. Multicollinearity Test
The purpose of multicollinearity testing is to test whether the regression model found a correlation between independent variables. If there is, it means there is multicolinearity. A good regression model should not have a correlation between independent variables (Santoso, 2001). Testing whether there is multi-colinearity can be detected by looking at the amount of VIF (Variance Inflation Factor) and tolerance value. The allowable VIF value limit is a maximum of 10. VIF values greater than 10 indicate high colinearity. Guidelines for a regression model that does not have multicollinearity are: VIF <= 10 1) Tolerance value> 0.1 where tolerance = 1 / VIF or VIF = 1 / tolerance

c. Autocorrelation Test
The autocorrelation test is a statistical analysis conducted to find out whether there is a correlation of variables that exist in a prediction model with a change in time. If the data is in a series of time series, autocorrelation test needs to be done. This test is important because a value in a particular observer sample is strongly influenced by the value of previous observations, whether in the linear regression model there is a correlation between the error of the intruder in period t and the error of the intruder in period t-1 (previous). In this study autocorrelation testing was performed using the Durbin-Watson method. In this case there are 2 values, namely du (durbin upper) and dl (durbin lower). If the value of durbin watson (d) is between du and 4 du, then there is no autocorrelation, and vice versa.
Requirements for autocorrelation among others: 1) If 0 <d <dl then there is no positive autocorrelation so the decision is rejected 2) If dl ≤ d ≤ du then there is no positive autocorrelation so there is no decision 3) If 4-dl <d <4 then there is no negative correlation so the decision is rejected. 4) If 4-du ≤ d ≤ 4 -dl then there is no negative correlation so there is no decision. 5) If du <d <4 -du then there is no positive or negative autocorrelation so the decision is not rejected. If the test using the Durbin Watson method is not successful, then it is carried out with a Run Test from the SPSS program, where if the Asymp value. Sig is greater than 0.05 then there is no autocorrelation.

d. Heteroskedacity Test
Heteroscedasticity test aims to test whether in the regression model there is an unequal variance from the residuals of one observation to another (Ghozali, 2016). If the variance from one residual to another observation residual is fixed, then it is called homocedasticity. If it is different, it is said that there is heterokedasticity. A good model is that if there is no heterokedastisitas in other words that if there is heterokedastisitas then the model is less efficient.
To detect the presence or absence of heteroscedasticity, the chart method (Scatterplot diagram) is used with the following provisions: 1) If there are certain patterns such as points, which exist to form a regular pattern (wavy, widened, then narrowed), then heteroscedasticity occurs. 2) If there is a clear pattern and the points spread above and below 0 on the Y axis, then heteroscedasticity does not occur.

Multiple Regression Analysis
In this study, the data analysis method used is multiple regression analysis method. Multiple regression analysis is used to determine whether there is a significant influence between the dependent variable when it is connected with two or more independent variables. In this study, the relationship of net profit margin, debt to equity ratio, and return on equity ratio on company value can be written as a multiple regression equation model as follows: = Return On Equity Ratio

Hypothesis Tests a. F-Test (Simultaneous Test)
This test is conducted to determine whether the regression model equation can be used. The F statistical test basically shows whether all the independent variables entered in the model have a simultaneous influence on the dependent variable. The steps in making a decision for the Fad test are: 1) If F count > F table and Sig. value <Α = 0.05 then it can be concluded that simultaneously / together, the independent variables significantly influence the dependent variable. 2) If F count <F table and Sig. value> Α = 0.05, it can be concluded that simultaneously / independent variables do not have a significant effect on the dependent variable. 3) Ho: b1 = 0, then the net profit margin, debt to equity ratio and return on equity ratio simultaneously / together, have no significant effect on company value / stock price. 4) Ha: b1 ≠ 0, then the net profit margin, debt to equity ratio and return on equity ratio simultaneously / together, have a significant effect on company value / stock prices.

b. t-test (Partial Test)
The statistical t test basically shows how far the influence of one independent variable individually in explaining the variation of the dependent variable (Ghozali, 2016).
The steps in making a decision for the t test are: 1) If t count > t table and Sig. value <α = 0.05, it can be concluded that partially the independent variable significantly influences the dependent variable. 2) If t count < t table and Sig. value > α = 0.05, it can be concluded that partially the independent variable has no significant effect on the dependent variable. 3) H b = b = 0 so the current ratio, debt to equity ratio and return on equity ratio have no significant effect on company value. 4) Hb ≠ b ≠ 0, then the current ratio, debt to equity ratio and return on equity ratio partially have a significant effect on company value.

c. The Coefficient of Determination (R 2 )
The coefficient of determination (R 2 ) essentially measures how far the model's ability to explain the variation of independent variables. The value of the coefficient of determination is between zero and one. A small R 2 value means that the ability of the independent variables to explain the variation of the dependent variable is very limited. A value close to 1 (one) means that the independent variables provide almost all the information needed to predict the dependent variable and vice versa if near zero does not provide the information needed.

IV. Discussion
The dependent variable in this study is company value (Price to Book Value). This data is obtained by distributing the value of Book Value per share with the share price. While the Book Value per share is obtained by sharing Total Equity with the number of shares outstanding. The independent variables used are net profit margin, debt to equity and return on equity. Here is the raw data or raw data from each variable. The variables in this study are described using descriptive statistics. The variables used in descriptive statistical analysis are net profit margin, debt to equity, return on equity, and price to book value. Descriptive statistical analysis is shown in Table 2 below:  Table 2, it can be seen that N or the total amount of data for each variable is 50 over the 2014-2018 period. The average value of net profit margin in this study was 0.128 times. The average value of debt to equity ratio in this study is 0.7956 times. The average value of return on equity in this study was 0.2576 times. The average price to book value in this study was 6.2994 times.
The independent variable Net profit margin in this study was 0.128 times. The average value of debt to equity ratio in this study was 0.7956 times. The average value of return on  The normality test aims to see whether in the regression model, confounding variables have a normal distribution. The initial normality test can be seen from the one-sample Kolmogorov-Smirnov Test table and the normal probability plot shown below. .002 c Data Sources are processed with SPSS 24.0 From the results of table 3 above, it can be seen that one-sample Kolmogorov-Smirnov table shows that the data are not normally distributed, because of the Asymp. Sig. value (2 tailed) obtained from NPM, DER, DER and PBV are 0.002, where the value of 0.002 is smaller than α = 0.05 indicating the results of data are not normally distributed.
Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www.  From the figure 2 above it can be seen that the data has always been left and right, indicating that the data is not normally distributed.   Based on the figure above, scatterplots graph after transformation shows that the points accumulate to the left of the graph, not spread either above or below the number 0 on the Y axis. It can be concluded that there is heteroscedasticity in the regression model. Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www. The results of this test give a significant Asymp. result of 0.2. This number is greater than 0.05, meaning that the data is normally distributed. If Asymp. Sig. (2-tailed) ≥ 0.05, the regression model produces normally distributed values. d. This is a lower bound of the true significance. Data Sources are processed with SPSS 24.0  The tolerance numbers are 0.285, 0.533, 0.250 where all are> 0.1 and the VIF number is 3.504, 1.8878, 4,000 where all are <10. These tolerance numbers indicate that this test passes the multicollinearity test because there is no multicollinearity between independent variables in the regression model. Based on the figure above, scatterplots graph after transformation shows that the points spread randomly and spread both above and below the number 0 on the Y axis. It can be concluded that there is no heteroscedasticity in the regression model. .086 Data Sources are processed with SPSS 24.0 Once again a run test was performed and the Asymp sig result was 0.086, already> 0.05, indicating the data passed the autocorrelation test.
Based on the SPSS output, we obtain the data to be used for the analysis of multiple linear regression equations as follows: Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www. bircu-journal.com/index.php/birci emails: birci.journal@gmail.com   The requirement for a relationship between the independent variables and the dependent variable is that t table <t counts in the calculation table above; 1) t count NPM: 0.080 <2.0086 (not related and not significant), means that NPM has no effect and is not significant to the company value, means that H1 is accepted 2) t count DER: 1,070 <2.0086 (not related and not significant), means that DER has no effect and on the company value, means that H2 is accepted 3) t count ROE: 8,246> 2,0086 (related and significant) means that DER has no effect and is not significant on company value, means that H3 is accepted Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www. Based on the results of the SPSS output it appears that the effect of simultaneously 3 independent variables (NPM, DER, and ROE) on the dependent variable (PBV) can be shown as in the following table. Seen from the results of test the above data obtained a significant value of 0,000. Significant value is smaller than 5% which can be concluded that the independent variable net profit margin, debt to equity ratio, return on equity together (simultaneously) affect the company value.
Also, the F-count results in the above test data (98.525) are greater than the F-table (2.79) so that it can be concluded that the net profit margin, debt to equity ratio, return on equity together (simultaneously) have a significant effect on the company value. The meaning of the determination coefficient test table is that NPM, DER, ROE variables can explain the company value at 85.7%. The remaining 14.3% was explained by other variables outside the study. The coefficient of determination test is used to test the goodness fit of the regression model. The number of 85.7 percent is close to 100 percent, means that the correlation between independent and dependent variables is very close.
The results of the study of the debt to equity ratio variable indicate that this variable has no effect on the company value and shows that the debt to equity ratio does not have an important role in increasing the company value. This means investors in making investment decisions are not based on DER indicators.
The company in determining its capital structure is not fully financed with debt. Even with the debt, the company will get savings on taxes, but the use of debt that is too large will burden the company's net profit in the case of paying a large interest expense. As far as greater benefits, additional debt is still permitted, but if the sacrifice due to the use of debt is already greater, then additional debt is not permitted. DER has no effect on company value. The size of the debt owned by the company is not given much attention by investors, because investors look more at how the company's management uses these effectively and efficiently to achieve added value to the company's value. This happens due to the fact that investors see that the debt incurred by the company can have a high or increasing cost of fund which will have an impact on the decline in corporate profits which will impact on the returns to investors.
Budapest International Research and Critics Institute-Journal (BIRCI-Journal) Volume 3, No 1, February 2020, Page: 494-510 e-ISSN: 2615-3076(Online), p-ISSN: 2615-1715 www. The results of this study is in line with the results of research by Alfredo Mahendra (2015), Alif Wahyu Wicaksana (2018) and Imam Ramantio's research (2018) namely that the debt to equity ratio has no effect on the company value in manufacturing companies listed on the IDX. But the results of this study not supported by the research of Dzulfikar Dwi Wahyu (2018) which states that DER influences the calculation of company value.
The results of the F Test and the coefficient of determination test illustrate that Net Profit Margin, Debt To Equity Ratio, and Return On Equity not only have a simultaneous effect on company value but also have a very large correlation, which is 85.7 percent in effecting company valuation.

V. Conclusion
Partially, Net Profit Margin has no effect on the company value in food and beverage companies on the Indonesia Stock Exchange in 2014-2018. Partially, Debt to Equity has no effect on the company value in food and beverage companies on the Indonesia Stock Exchange in 2014-2018. Partially, Return on Equity affects the company value in food and beverage companies on the Indonesia Stock Exchange in 2014-2018. Net Profit Margin, debt to equity ratio, return on equity simultaneously have a significant effect on company value in food and beverage companies on the Indonesia Stock Exchange in 2014-2018.