Forecasting refers to the process of making informed estimates based on historical data. It can be seen as a planning technique that allows people to strategize their next moves and that can be done qualitatively or quantitatively.
Qualitative forecasting is based on the opinion and judgment of experts or consumers and is generally used when quantitative data is not available. Examples of these techniques are the Delphi method, market research, and historical life-cycle analogy.
Quantitative methods utilize numerical data and statistical methods to draw information. Typical methods include time series analysis, analysis of indices or indicators, econometric modeling, machine learning modeling, and deep learning modeling.
The data used in all these methods can come from two different types of sources: primary and secondary. Primary sources are first-hand data, usually obtained by the researcher through surveys, measuring instruments, market data collections, and more. Differently, secondary sources consist of information obtained and interpreted by others that is used to draw conclusions.
Forecasting is one of the most important tools in science and engineering. In physics and chemistry, it is used to verify theories and ideas by predicting phenomena and checking their occurrence. In finance, it is one of the most fundamental tools for investors, who use forecasting to analyze different investment options and decide on a strategy. And in economics, it is used to estimate future inflation, GDP, and other economic metrics that provide a view of how a country’s economy is performing.
Research has shown that different models lead to different prediction accuracies. LogicPlum’s platform solves this problem through automation, allowing researchers and investors alike to consider hundreds of different algorithms and select the optimal one quickly. As a consequence, its users don’t need proficiency in mathematics or statistics but only in their field of knowledge, such as marketing, finance, economics, or engineering.
For those interesting in this subject, a very good article from the Harvard Business Review is
Chambers, J. et al. (1971). How to Choose the Right Forecasting Technique. Available at https://hbr.org/1971/07/how-to-choose-the-right-forecasting-technique
© 2020 LogicPlum, Inc. All rights reserved.