|
Keywords
|
Inventory management efficiency, Predictive variables, CEO,
Artificial Neural Networks, Back propagation of Error algorithm.
|
|
Abstract
|
The purpose of this study is to design a model to predict the efficiency of
inventory management to help creditors and actual and potential investors and
other stakeholders to avoid major losses in the capital market. For this reason,
137 companies listed on the Tehran Stock Exchange during the 10-years period
2012-2021 were examined. In this study, the predicting variables of institutional
ownership, managerial ownership, corporate ownership, ownership
concentration, board size, percentage of non-executive board members, and
duality of CEO (Chief Executive Officer) role have been used. The efficiency of
inventory management was predicted using a three-layer perceptron artificial
neural network with the Back propagation of Error algorithm. Finally, a network
with the mean squared error of 0.360, 0.428, 0.261 and 0.353, respectively for
training data, validation, test and total data and a coefficient of determination of
more than 72%, as the best network Selected.
|