The manufacturing company is the main pillar of industrial development in a country. The development of manufacturing industry can be used as a benchmark to see national industrial development in the country. The performance of a good manufacturing company can be seen from profitability. Profitability is the ability of companies to earn profits in relation to sales, total assets capital. Profitability ratios are measured by Net Profit Margin (NPM). This ratio measures the ability of a company to generate profitability at a certain level of sales, assets, and capital stock. The greater the Net Profit Margin, the more efficient the company is in issuing the costs associated with its operations. The relationship between profitability and the factors that influence it will be studied to obtain a mathematical model. This mathematical model will show the factors that influence profitability significantly. Predictor variables which determine profitability are Leverage, Manufacturers' Size, Liquidity, and Tangibility. In conducting research on profitability, the data used is a combination of cross section and time series or panel data. This data is used because it is necessary to observe the behavior of research units at various time periods. This study used secondary data of 2008 - 2016 Annual Reports which were downloaded from 21 related manufacturers' official websites. One of the statistical analysis tools used, to observe the behavior of research units at various time periods, is ordered logistic regression analysis in panel data, which is an extension of logistic regression when it is used in panel data. Estimation of binary logistic model parameters in panel data using maximum likelihood estimation method with Gauss-Hermitte Quadrature iteration. Based on the best model obtained, the factors that influence manufacturers' profitability, in Indonesia, are leverage and tangibility. The result of the Likelihood Ratio Test shows that the random effect panel binary logistic regression model is better model than standard binary logistic regression with a classification of accuracy of 73.54%.
|Number of pages||18|
|Journal||International Journal of Innovation, Creativity and Change|
|Publication status||Published - 2019|
- Binary Logistic Regression
- Panel Data
- Random Effect