Forecasting; the Power of Scientific Prediction and Calculation

To understand in a simple manner, forecasting is simply a prediction of what will happen in the near future.  Understanding the word “prediction” it implies a variable movement of events in the near upcoming occurrence.  The higher the chances of the event to happen the closer the given dependent variable to overlap with the independent variable, meaning the given prediction has a higher chance to happen in the near future.  

Some important topics under forecasting that must be noted are the topic of Moving Averages and Exponential Smoothing, Trend Projection, Seasonality and Trend, and Time Series and decomposition.  Moreover, patterns used in time series must be observed to classify and differentiate the different statistical cases from the real business events happening in the real world.  Using a wrong forecasting tool will affect the output data of the case which will eventually lead to the miss calculation and planning. 

Accurateness in digesting the raw data through statistical tools will reveal the right informative forecasting.  Undeniably, we sometimes argue that data results are not true because the calculation process and conceptualization of the data is not accurate.  Therefore, in statistical data analysis accurate statistical practice must be observed to avoid erroneous data output.  Another aspect that a researcher must consider is the quantity of respondents who will participate the research data gathering.  

The basic rule of research data gathering is, the larger the population involved in the data gathering the higher the chances of accuracy.  Generally, researchers try to gather as much data from the respondents to cover the required number of samples.

ASPECTS TO BE CONSIDERED:

Moving Average and Exponential Smoothing- is a technique of predicting the next future event based from the latest observation.  Therefore, every occurrence of the observed variable will change the future behavior of the variable.  Furthermore, accumulating the behavioral events in a certain period of time will create a time series which will be more understandable by looking in graph presentation.

Weighted Moving Averages- is selecting a different weight for each data value and later computing a weighted average of the latest k values as the forecast.

Mean absolute error (MAE)- The average of the absolute values of the forecast errors.

Mean squared error (MSE)- the average of the sum of squared forecast errors.

Mean absolute percentage error (MAPE)- The average of the absolute values of the percentage forecast errors.

PRACTICAL BUSINESS APPLICATION SAMPLES:

Forecasting for food and beverage sales- by using the forecasting tool the management of the sample company can calculate the volume of supply that the company will deliver to the market as well as the product volume considering the seasonal fluctuation of demand.

Lost Sales- implies to the specific monitoring of the lost sales of the company or more specifically to a given product line daily sales loss.