Quantitative analysis, the use of statistical modeling for the purpose of providing a predictive capability, is often used in sales forecasting for the logical reason that it exists largely for that purpose. By accumulating data on past and current commercial transactions and examining that data for patterns, sales managers and other corporate executives are able to predict future sales activities. While statistical modeling is used for projecting sales, it is not without flaws. Product development is predicated upon current activity and careful assessments of future market trends. In other words, today's product may not be in demand tomorrow. Consumer tastes change, and wholesalers and retailers alike loathe the prospect of ending up with large surpluses or inventories of items no longer in demand. Such surpluses represent wasted resources, including the space occupied by those items, the costs of the raw materials involved in their manufacture, and associated shipping costs. Statistical modeling will always play a role in sales forecasting. It is not, however, an fallible process for projecting into the future. Consumer demand, often dependent upon both current trends and economic considerations, is too fickle. Statistical modeling does not reflect either of these variables, so it is inherently limiting. It remains an important component in forecasting sales, but it is not without its risks.

Statistical planning entails the collection of data and the analysis of this information. Statistical planning is a direct result of statistical inference, and statistical inference is basically the interpretation of any given data set. In inferential statistics, the analysis is geared toward the larger population of any given data set, and after the properties of this given data set are interpreted, this information is then used to form and test hypotheses that inform subsequent estimates based upon data sampled from a larger population. In sales forecasting, statistical planning would be necessary to form a comprehensive understanding of market supply and demand. By using data from a given population, businesses can better understand the behavior of consumers, and with this understanding, they can develop business models that are tailored to market demand. Statistical planning is just the mode by which businesses collect and interpret data. This information is then used to inform sales forecasting, which is the process of estimating sales within a specific market. Sales forecasting protects businesses from profitless expenditures, and businesses use both of these practices in order to protect their investments and capitalize on economic trends.

I believe you want to know about statistical methods of sales forecasting.

Sales forecasting, also called demand forecasting is job of estimating or forecasting what is likely to be demand for a product in a specified period in future. Sales can be forecast using several methods. In addition to statistical methods, these includes other methods such as qualitative judgement, order book, and explosion of bill of material.

Statistical forecasting itself consists of many different types of techniques such as time series, causal (regression analysis), market research, and simulation. The time series used data of demand in the past to estimate the pattern of demand in terms of variables such as average demand, trend, and seasonal variability to forecast demand in future.

Causal forecasting uses regression analysis to determine the relationship of demand of a product with some other known variable and uses that to forecast demand. For example it may be possible to determine relationship between weather and sale of ice-cream. Using this understanding sale of ice-cream may be forecast based on expected temperature.

Market research is based on collecting sample data by direct observation or by some other method such as interviewing people, and then statistically extrapolating the future demand from such data.

Simulation is used to forecast demand when the demand is dependent on many different interrelated variables and it is not possible to create an integrated model of demand forecasting that results in one single forecast. The simulation method consist of describing the relationships of demand with relevant influencing factors in term of statistical probabilities and based on that working out likely demand under different scenarios. The Insights obtained for these results of simulation studies are used of deciding on the most likely value for the expected demand.