A method for forecasting drug sales volume

Abstract

The invention discloses a method for forecasting drug sales volume, which comprises the following steps: S1, establishing a basic information relationship: firstly, cleaning and arranging data, collecting bill information of all hospitals in a region, and constructing a tree information map of patients, diseases of patients in the past period, and drugs and medical devices used by patients; S2, calculating the characteristic index of the patient according to the list of the patient, and applying the canopy algorithm and the KMeans algorithm to carry on the machine learning, and classifying thepatient automatically; S3,using an ARIMA model of time series analysis to predict the number of patients of each type in the next year; S4, calculating the proportion of each disease of the drug or medical device, and obtaining the predicted value of the drug or medical device based on the disease composition according to the predicted value of each type of patient. The method has the advantagesof reasonable design, simple and reliable prediction mode, and convenient and accurate prediction of the sales volume of various medicines.

Description

technical field [0001] The invention relates to the technical field of sales forecast, in particular to a method for medicine sales forecast. Background technique [0002] At present, most of the predictions of disease incidence and drug sales are based on time series analysis methods. The time series analysis method uses past historical data and further speculates on future development trends through statistical analysis. It is based on the fact that all things are developing and changing, and the development and changes of things have continuity in time. [0003] The time series forecasting method only considers the time factor and does not consider the influence of external factors on the forecasted object. When there is a large change in the outside world, there will often be a large deviation. Especially for the incidence of diseases and the sales of medicines, which are greatly affected by external factors such as the environment, when making predictions, only conside...

AI Technical Summary

Problems solved by technology

[0003] The time series forecasting method only considers the time factor and does not consider the influence of external factors on the forecast object. When there is a large change in the outside world, there will often be a large deviation

Especially for the incidence of diseases and the sales of medicines, which are greatly affected by external factors such as the environment, when making predictions, only considering the time factor, the prediction results will not match the actual situation

Images