Sample store forecasting process and system
View Patent ↗A method and system of predicting market information includes the steps of receiving first data, forecasting further data based on the first data, receiving second data and comparing the further data with the second data, and creating an adjustment factor to account for any difference between the further data and the second data.
1. A method for forecasting un-reported prescription/product transactions or transactions that are not timely reported in a subject time interval to a sample store or outlet in a universe of product stores, the universe of product stores comprising sample stores and non-sample stores in market channels such as retail, mail order, and long term care, the sample stores generally reporting prescription transaction data to a history database, the method comprising the steps of:
using a computer processor to:
identify new products that have been launched in a number of recent weeks using the sales volume change of the products in the number of recent weeks in the prescription transactions stored in the history database;
assign products from the prescription transaction data to product groups based on sales volume of the products or therapy class;
for each product group,
generate data files including projected national prescription count information by product for each of a number of channels, and
generate data files including raw prescription counts at the outlet/product level;
combine the datafiles including projected national prescription count information and raw prescription counts;
identify outlets as normal volume or low volume outlets using a predefined sales volume threshold, wherein the low volume outlets are identified as the outlets having a sales volume smaller than the predefined sales volume threshold, and the normal volume outlets are identified as the outlets having a sales volume equal or greater than the predefined sales volume threshold;
for low volume outlets, using a 4 week average of outlet/product raw prescription counts as a forecast for the current week volume;
for normal volume outlets, when the product is not new, using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume based on outlet/product raw prescription counts for a prior number of weeks and the generated projected national prescription counts for both the current week and the prior number of weeks;
for normal volume outlets, when the product is new, using a national ratio of product prescription counts to product group prescription counts applied at outlet level to forecast a new product volume for the current week; and
generate a combined forecast based at least in part on the forecasts in connection with the low and normal volume outlets.
2. The method of claim 1 , wherein using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume comprises using an Autoregressive Integrated Moving Average (ARIMA) model.
3. The method of claim 1 , wherein identifying new products that have been launched in a number of recent weeks comprises identifying new products that have been launched in a the last thirteen weeks and supplementing the data files with historical raw prescription data at the outlet/product level covering a prior number of weeks comprises raw prescription data for twenty five prior weeks.
4. The method of claim 1 , wherein assigning products to product groups comprises identifying a number of leading products based on analysis of national prescription counts information and assigning such number of products to its own product group.
5. The method of claim 1 , wherein assigning products to product groups comprises grouping products by therapy class.
6. The method of claim 1 , wherein identifying outlets as normal volume or low volume outlets is based at least in part on average prescriptions per week, number of missing weeks, or maximum prescriptions per week.
7. A system for forecasting un-reported prescription/product transactions or transactions that are not timely reported in a subject time interval to a sample store or outlet in a universe of product stores, the universe of product stores comprising sample stores and non-sample stores in market channels such as retail, mail order, and long term care, the sample stores generally reporting prescription transaction data to a history database, the system comprising:
a computer processor and computer readable storage media;
the computer processor configured to perform steps comprising:
identifying new products that have been launched in a number of recent weeks using the sales volume change of the products in the number of recent weeks in the prescription transactions stored in the history database;
assigning products from the prescription transaction data to product groups based on sales volume of the products or therapy class;
for each product group,
generating data files including projected national prescription count information by product for each of a number of channels, and
generating data files including raw prescription counts at the outlet/product level;
combining the datafiles including projected national prescription count information and raw prescription counts;
identifying outlets as normal volume or low volume outlets using a predefined sales volume threshold, wherein the low volume outlets are identified as the outlets having a sales volume smaller than the predefined sales volume threshold, and the normal volume outlets are identified as the outlets having a sales volume equal or greater than the predefined sales volume threshold;
for low volume outlets, using a 4 week average of outlet/product raw prescription counts as a forecast for the current week volume;
for normal volume outlets, when the product is not new, using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume based on outlet/product raw prescription counts for a prior number of weeks and the generated projected national prescription counts for both the current week and the prior number of weeks;
for normal volume outlets, when the product is new, using a national ratio of product prescription counts to product group prescription counts applied at outlet level to forecast a new product volume for the current week; and
generating a combined forecast based at least in part on the forecasts in connection with the low and normal volume outlets.
8. The system of claim 7 , wherein using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume comprises using an Autoregressive Integrated Moving Average (ARIMA) model.
9. The system of claim 7 , wherein identifying new products that have been launched in a number of recent weeks comprises identifying new products that have been launched in a the last thirteen weeks and supplementing the data files with historical raw prescription data at the outlet/product level covering a prior number of weeks comprises raw prescription data for twenty five prior weeks.
10. The system of claim 9 , wherein assigning products to product groups comprises identifying a number of leading products based on analysis of national prescription counts information and assigning such number of products to its own product group.
11. The system of claim 9 , wherein assigning products to product groups comprises grouping products by therapy class.
12. The system of claim 9 , wherein identifying outlets as normal volume or low volume outlets is based at least in part on average prescriptions per week, number of missing weeks, or maximum prescriptions per week.
13. A non-transitory computer-readable medium for forecasting un-reported prescription/product transactions or transactions that are not timely reported in a subject time interval to a sample store or outlet in a universe of product stores, the universe of product stores comprising sample stores and non-sample stores in market channels such as retail, mail order, and long term care, the sample stores generally reporting prescription transaction data to a history database, the computer-readable medium having a set of instructions operable to direct a processing system to perform the steps of:
identifying new products that have been launched in a number of recent weeks using the sales volume change of the products in the number of recent weeks in the prescription transactions stored in the history database;
assigning products from the prescription transaction data to product groups based on sales volume of the products or therapy class;
for each product group,
generating data files including projected national prescription count information by product for each of a number of channels, and
generating data files including raw prescription counts at the outlet/product level;
combining the datafiles including projected national prescription count information and raw prescription counts;
identifying outlets as normal volume or low volume outlets using a predefined sales volume threshold, wherein the low volume outlets are identified as the outlets having a sales volume smaller than the predefined sales volume threshold, and the normal volume outlets are identified as the outlets having a sales volume equal or greater than the predefined sales volume threshold;
for low volume outlets, using a 4 week average of outlet/product raw prescription counts as a forecast for the current week volume;
for normal volume outlets, when the product is not new, using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume based on outlet/product raw prescription counts for a prior number of weeks and the generated projected national prescription counts for both the current week and the prior number of weeks;
for normal volume outlets, when the product is new, using a national ratio of product prescription counts to product group prescription counts applied at outlet level to forecast a new product volume for the current week; and
generating a combined forecast based at least in part on the forecasts in connection with the low and normal volume outlets.
14. The computer readable medium of claim 13 , wherein using a moving four-week average of outlet/product raw prescription counts to forecast the current week volume comprises using an Autoregressive Integrated Moving Average (ARIMA) model.
15. The computer readable medium of claim 13 , wherein identifying new products that have been launched in a number of recent weeks comprises identifying new products that have been launched in a the last thirteen weeks and supplementing the data files with historical raw prescription data at the outlet/product level covering a prior number of weeks comprises raw prescription data for twenty five prior weeks.
16. The computer readable medium of claim 13 , wherein assigning products to product groups comprises identifying a number of leading products based on analysis of national prescription counts information and assigning such number of products to its own product group.
17. The computer readable medium of claim 13 , wherein assigning products to product groups comprises grouping products by therapy class.
18. The computer readable medium of claim 13 , wherein identifying outlets as normal volume or low volume outlets is based at least in part on average prescriptions per week, number of missing weeks, or maximum prescriptions per week.