Improvement of Prosperity Prediction in Czech Manufacturing Industries

Michal Kubenka

Abstract


There are many ways to determine financial health of a company. Classic tools of financial analysis emphasize data in financial statements. When trying to reach an overall classification of a company, e.g. from the point of view of an investor, indicators of profitability, liquidity, leverage and activity might generate contradictory clues regarding financial health of the company. Such analysis does not only generate ambiguous results, but is also lengthy. Therefore, the IN05 financial model was created for particular specifics of Czech Republic economic environment to predict prosperity of companies. It measures the prosperity on the base of economic value added (EVA), in that this model is globally unique. More than ten years have passed since the launch of the IN05 index and the Czech Republic and the world have in the meantime weathered a global economic crisis. The aim of this study is to quantify contemporary prediction strength of the IN05 index on a sample of 500 companies. Moreover, an effort was made to enhance the accuracy of the model for particular branches of manufacturing industries. It was found that contemporary explanatory power of the index IN05 is 48.58% when applying the original evaluation scale, and 70.79% when assessing 68.62% of the sample except the gray zone. The accuracy of the model was enhanced by author in all categories, i.e. a higher accuracy (correct prediction probability) without considering the gray zone and higher accuracy (conditioned correct prediction probability) using the gray zone in all branches of manufacturing industries. Second aim is confirmation of branch influence on accuracy of prosperity prediction. It was found a different IN05 model accuracy for selected branches with statistical significance.

DOI: http://dx.doi.org/10.5755/j01.ee.29.5.18231


Keywords


prediction model; financial health; IN05 index; economic value added; return on equity; manufacturing industry

Full Text: PDF

Print ISSN: 1392-2785
Online ISSN: 2029-5839