# Projects

## REGRESSION MODELING OF PROFIT FOR PHARMACEUTICAL INDUSTRY

The industries aim at profit. To Maximize the profit is dream of every Industrialist. Here an attempt is made to formulate profit function for pharmaceutical industry.
Regression analysis, Buckingham-pi theorem & artificial neural network are generally used for this purpose.

This work developed regression equation by using four input parameters. Many
financial inputs have their own effect on profit. Initially twenty such parameters were selected for forty different pharmaceutical companies. The parameters were selected on the basis of their availability for all companies.

The different input parameters were Equity share capital, Preferential Share Capital, Reserve & Surplus, Secured Loans, Unsecured Loans, Gross Block, Accumulated Depreciation, Capital Work in Progress, Investment, Current Liabilities, Net Operating Income, Material Consumption, Manufacturing
expenses, Personnel Expenses, Selling Expenses, Administrative Expenses, Financial Expenses, Depreciation, Tax Changes, Equity Dividend. The output parameter was selected as net profit.

By visual inspection four parameters viz. Preferential Share Capital, Accumulated Depreciation, Tax Changes and Equity Dividend were dropped. Considering the interdependence of remaining parameters some of them were eliminated and finally four factors were used.

Elimination of inputs is done using correlation matrix.
Procedure of selection of very important variables is similar to Vogel’s approximation method. It is an effective application of the "best cell method" along with some tie-breaking features.

The matrix is obtained by adding the "row opportunity cost Matrix" (row opportunity cost matrix: for each row, the smallest cost of that row is subtracted from each element of the same row) and the "column opportunity cost matrix"
A simple linear function of input parameters is developed using regression analysis to represent profit.

Behavior of profit can now be understood with an ease. Error is an inherent part of this process: it provides near approximation. Still the extent of influence of each input is clearly understood. Coefficients of Input variable are the indicator of its influence on output.

This model provides the direction for managing the Industry. Some times it is not possible to increase the single Input and some more Inputs are to be adjusted simultaneously. The clear strategies for progress can now be devised. A similar kind of analysis for other sectors can also be done.

Tags :

## Posted by

Tue, 12/04/2011 - 07:57