Question

Your responses are to be posted as well thought out scholarly paragraphs that convey additional meaningful insights.

• How are predictive models useful for forecasting purpose?
• How can you encourage your company to implement the use of predictive models?

Answer 1

Predictive modeling is a process of using known statistics,data mining and probablity to validate a model that can be used to forecast future outcomes. It is also known as predictive analytics,predictive analysis or machine learning. These models can be used to predict anything like the price of real estate,ratings of tv shows etc. These models may be made up of simple linear equations or of complex neural networks.

predictive modeling can be used for whether forecasting,classifying mails,etc. The data's used for prediction purposes are Transaction data,Customer service data,Digital marketing and advertising data,Demographic data,Geographical data,CRM data,Survey or polling data,Economic data,Machine-generated data,Web traffic data,etc.

There are different types of prediction models they are linear regression,logistic regression,decision trees and time series analysis. Logistic regression uses only binary values like 0 or 1. Linear regression finds a linear relationship between independent variable and dependent variable. Decision tree is used for classification in which each node represents the attributes and each branch represents the conditions.

Different applications are the following:
1.Algorithmic trading-used for stock,commodity,currency trading,etc.
2.Health care-used in prediction of different diseases like cancer,etc.
3.Auto insurance-Insurance policies are being made by predicting the risk of accident based on past history of a location.
4.Customer relationships management-like predicting which all actions can lead to a purchase,etc.

Limitations
1.Accurate prediction of future is not always possible using the history data.
2.Some attributes may not be considered during data collection.
3.adverse effects can occur when people learn the algorithm and manipulate it.

Predictive models should be used in any companies since it clarifies our objectives,helps in building our infrastructure,helps in defining success and data maintenance can be done effectively.