ZY Gift Company’s Cost Forecasting Study

The actual problems such as inventory backlog,stagnant market,difficulty in recovering payments,and declining employee benefits not only make the survival and development of the gift company face serious challenges,but also put forward stricter and higher standard requirements for the company’s cost management.Strict control of costs is the main means for enterprises to reduce costs and increase efficiency.In order to effectively monitor the cost expenditure situation in real time,adjust the cost strategy in time and enhance the core competitiveness of enterprises.In this paper,the cost system of ZY Gift Company is studied by using expert survey method,literature research method,and combined qualitative and quantitative analysis.Focusing on cost accounting methods and cost forecasting,we establish an order costing system under the operation cost method,and find out the main factors affecting the order cost,and use algorithms to forecast the company’s order cost to achieve effective and timely control of the cost.Firstly,the existing order cost sharing method of ZY Gift Company was improved.After analyzing the traditional cost accounting method,the operation cost method is applied to ZY gift company’s order costing by combining ZY gift company’s own operation characteristics,and the accounting cost is determined separately by dividing the accounting objects into operations.Secondly,the main indicators affecting the order cost of ZY gift company were analyzed from four aspects: labor factor,sales factor,purchasing factor and logistics factor.Following that,the principal component analysis theory was applied to reduce the dimensionality of the collected cost data of ZY company,and six principal components with eigenvalues greater than 1 were extracted.Then,the principal component scores were calculated to derive the input data for the subsequent prediction model.Finally,a long and short-term memory network was used to construct an order cost prediction model for ZY gift company,and the input data processed by principal component analysis was divided into a training set and a test set,and the test sample data was predicted after a better prediction effect was obtained.And the prediction results are compared with the effect of traditional cost budgeting method,GM(1,1)cost prediction model,and LSTM to predict the order cost directly.Through the study,the existing order cost accounting system of ZY Gift Company was improved,which is more conducive to its cost control.By comparing the effect with other forecasting methods,it is found that after the principal component analysis of the original index,the feasibility of the index is improved,the workload of the forecasting model is effectively reduced,the running time is shortened,the correlation between the predicted and actual values of the order cost obtained by forecasting is higher,and the forecasting model has a better prediction effect.It is conducive to accurately predicting the future order cost of ZY Gift Company,so that the enterprise can have a basis for formulating cost budget plan and a justification in cost management.