/
Forecast Analysis Components

Forecast Analysis Components

Forecast analysis components (the “Components” element) are intended for detailed analysis and examination of the constructed forecasting model.

Forecast analysis components are divided into the following categories:

  • Units – components that display forecast results in sales volumes (units);

  • Monetary – components that display historical data and forecast results in monetary terms;

  • Historical – components that display historical data and intermediate forecast results (predictors used at any stage of forecasting), calculated based on historical data;

  • Competitors – components that display competitor prices for the selected item;

  • Regressors – components that display the impact of price factors, weather factors, and assortment width within the group;

  • Economic Effect – components that display the economic effect of proper use of MySales;

  • Models – components that display forecast results using alternative forecasting models (not displayed for aggregated models).

Category: Units

The Units category consists of the following components:

  • Order – displays forecasted sales volume in units including safety stock (recommended order result);

  • Seasonality – displays seasonal sales growth in units (considering the forecast impact coefficient);

  • Trend – displays the impact of trend on sales volumes in units (considering the forecast impact coefficient);

  • Region – displays the impact of sales redistribution by region (considering the forecast impact coefficient);

  • Group – displays the impact of sales redistribution by group (considering the forecast impact coefficient);

  • Auto Error – displays the restored influence of other factors (considering the forecast impact coefficient). Calculated based on the Average Error component. The duration may be 2, 4, or 6 weeks depending on which duration provides the best result (lowest error);

  • Min – displays the transformation impact if the forecast is below the historically acceptable minimum sales volume in units (considering the forecast impact coefficient). By default, the lower quartile of sales volume is considered the minimum acceptable threshold;

  • Price % – displays the impact of price changes on sales volumes in units (considering the forecast impact coefficient);

  • Discount – displays the impact of discount in monetary terms on sales volumes in units (considering the forecast impact coefficient);

  • Discount % – displays the impact of percentage discount on sales volumes in units (considering the forecast impact coefficient);

  • Exchange Rate – displays the impact of the USD exchange rate on sales volumes in units (considering the forecast impact coefficient);

  • Actual Price – displays the impact of the real price (price including discount) on sales volumes in units (without applying the forecast impact coefficient);

  • Price – displays the impact of price on sales volumes in units (considering the forecast impact coefficient);

  • Assortment Width – displays the impact of available assortment width (assortment width in stock) on sales volumes in units (considering the forecast impact coefficient);

  • Group Discount – displays the impact of the average discount for all SKUs in the group to which the item belongs (considering the forecast impact coefficient);

  • Group Price – displays the impact of the average price for all SKUs in the group to which the item belongs (considering the forecast impact coefficient);

  • Base – displays the impact of the base sales level (average sales) on forecasted sales volumes in units (considering the forecast impact coefficient);

  • Temperature – displays the impact of average weekly air temperature on sales volumes in units (considering the forecast impact coefficient);

  • Rain – displays the impact of the share of rainy days during the week on sales volumes in units (considering the forecast impact coefficient);

  • Snow – displays the impact of the share of snowy days during the week on sales volumes in units (considering the forecast impact coefficient);

  • Anomalies – displays anomalous sales volume in units and allows manual input in the opened field. Anomalies are described in detail in the section “Anomalies and Exceptions”;

  • Exceptions – displays excluded weeks and allows excluding weeks in the opened field. Exceptions are described in detail in the section “Anomalies and Exceptions”.

If some components from the list are not displayed, this means they were not included in the forecasting model as predictors.

The ® symbol or the prefix “reg” before a component means that this component is included in the restored value (regressors).

Category: Monetary

The Monetary category consists of the following components:

  • Sales – historical sales volume in monetary terms;

  • Forecast – forecasted sales volume in monetary terms;

  • Past Sales – historical sales volume in monetary terms shifted 52 weeks forward in time;

  • Order – displays forecasted sales volume in monetary terms including safety stock (recommended order result);

  • Seasonality – displays seasonal sales growth in monetary terms (considering the forecast impact coefficient);

  • Trend – displays the impact of trend on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Region – displays the impact of sales redistribution in monetary terms by region (considering the forecast impact coefficient);

  • Group – displays the impact of sales redistribution in units by group (considering the forecast impact coefficient);

  • Auto Error – displays the restored influence of other factors in monetary terms (considering the forecast impact coefficient). Calculated based on the Average Error component. The duration may be 2, 4, or 6 weeks depending on which duration provides the best result (lowest error);

  • Min – displays the transformation impact if the forecast is below the historically acceptable minimum sales volume in monetary terms (considering the forecast impact coefficient). By default, the lower quartile of sales volume is considered the minimum acceptable threshold;

  • Price % – displays the impact of price changes on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Discount – displays the impact of discount in monetary terms on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Discount % – displays the impact of percentage discount on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Exchange Rate – displays the impact of the USD exchange rate on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Actual Price – displays the impact of the real price (price including discount) on sales volumes in monetary terms (without applying the forecast impact coefficient);

  • Price – displays the impact of price on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Assortment Width – displays the impact of available assortment width (assortment width in stock) on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Group Discount – displays the impact of the average discount for all SKUs in the group to which the item belongs (considering the forecast impact coefficient);

  • Group Price – displays the impact of the average price for all SKUs in the group to which the item belongs (considering the forecast impact coefficient);

  • Base – displays the impact of the base sales level on forecasted sales volumes in monetary terms (considering the forecast impact coefficient);

  • Temperature – displays the impact of average weekly air temperature on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Rain – displays the impact of the share of rainy days during the week on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Snow – displays the impact of the share of snowy days during the week on sales volumes in monetary terms (considering the forecast impact coefficient);

  • Anomalies – displays anomalous sales volume in monetary terms and allows manual input in the opened field.

If some components from the list are not displayed, this means they were not included in the forecasting model as predictors.

The ® symbol or the prefix “reg” before a component means that this component is included in the restored value (regressors).

Category: Historical

The Historical category consists of the following components:

  • Seasonality – displays seasonal sales growth;

  • Trend – displays the impact of trend on sales volumes;

  • Actual Price – displays the real price (price including discount);

  • Price – displays price;

  • Price % – displays how many times the price has changed;

  • Discount – displays discount in monetary terms;

  • Discount % – displays discount in percentage;

  • Stock – displays stock balance in the warehouse;

  • Group Price – displays the average price for all SKUs in the group to which the item belongs;

  • Group Discount – displays the average discount for all SKUs in the group to which the item belongs;

  • Assortment Width – displays available assortment width (assortment width in stock);

  • Region – displays the impact of sales redistribution by region;

  • Group – displays the impact of sales redistribution by group;

  • Exchange Rate – displays the USD exchange rate;

  • Exchange Rate % – displays how many times the USD exchange rate has changed;

  • Seasonal Growth – displays seasonal sales growth as a coefficient (how many times sales will increase);

  • Availability – displays product availability (whether stock was sufficient for maximum sales). When displayed per store: “0” or blank means unavailable, “1” means sufficient stock. When displayed across multiple stores, shows the number of stores with sufficient stock;

  • Temperature – displays average weekly air temperature;

  • Temperature Δ – displays the difference between the average temperature of a specific week and the average temperature over the last 13 weeks;

  • Rain – displays the share of rainy days during the week;

  • Snow – displays the share of snowy days during the week;

  • Season Type – displays season type (0 = regular season, -1 = low season, 1 = high season);

  • Promo Coefficient – displays the share of promo days in a week as a percentage;

  • Average Error – displays average forecast error;

  • Auto Error – displays restored influence of other factors. Calculated based on the Average Error component. The duration may be 2, 4, or 6 weeks depending on which duration provides the best result (lowest error);

  • Median – displays the median sales value (a value such that exactly half of the observations are above it and half are below it);

  • Base – displays the base sales level;

  • Min – displays transformation of sales volume if the forecast is below the historically acceptable minimum threshold. By default, the lower quartile of sales volume is considered the minimum threshold;

  • Inbound – displays the total amount of goods received by stores (from supplier or warehouse), minus write-offs and shortages, including surplus postings;

  • # Stores – displays the number of stores where the group or SKU is represented (available at region level).

Category: Regressors

The Regressors category consists of the following components:

  • Reg. Actual Price – displays the impact of real price (price including discount) on sales volumes;

  • Reg. Price – displays the impact of price on sales volumes;

  • Reg. Price % – displays the impact of price change on sales volumes;

  • Reg. Discount – displays the impact of discount in monetary terms on sales volumes;

  • Reg. Discount % – displays the impact of percentage discount on sales volumes;

  • Reg. Exchange Rate – displays the impact of the USD exchange rate on sales volumes;

  • Reg. Exchange Rate % – displays the impact of USD exchange rate change on sales volumes;

  • Reg. Assortment Width – displays the impact of available assortment width (assortment width in stock) on sales volumes;

  • Reg. Temperature – displays the impact of average weekly air temperature on sales volumes;

  • Reg. Rain – displays the impact of the share of rainy days during the week on sales volumes;

  • Reg. Snow – displays the impact of the share of snowy days during the week on sales volumes.

Category: Economic Effect

The Economic Effect category consists of the following components:

  • Stock Reduction – shows how much forecast usage can reduce warehouse stock levels;

  • Lost Sales – shows lost sales due to an underestimated forecast;

  • Sales Increase – shows how much forecast usage can increase sales.