Discuss the following concepts of Module 2
Discuss how the coefficient of determination and the coefficient of correlation are related and how they are used in regression analysis.
Discuss the problems and drawbacks of the moving average forecasting model.
What effect does the value of the smoothing constant have on the weight given to the past forecast and the past observed value?
Your participation in the discussions counts toward the overall assessment for this course. You are required to post at least one original message and to respond to two or more of the other students’ postings. Make sure you support your points using concepts and methods of Module 2 and more so the content of the chapters in this module. Write clearly and precisely. Postings like, I agree or disagree, yes, no etc. do not count.
student respond 1
1. Coefficient of determination and the coefficient of correlation
Regression analysis is usually carried out to show how variables relate or affect each other. In this analysis, there is usually an outcome variable and predictor variable (s). the relationship between the variables is explained through the correlation coefficient. The correlation coefficient is employed to show the direction and strength of the relationship between the variables of a model. On the other side, how well the regression model represents the data is explained through the coefficient of determination (Piepho, 2019). That is, the coefficient of determination will explain the variation arising within the model. Therefore, the coefficient of determination and correlation coefficient enables a researcher to understand the data and how the data behaves.
2. Problems and drawbacks of the moving average forecasting model
The models are usually employed in time series analysis, and their application is due to their flexibility and simplicity when analyzing data. However, these models have their drawbacks, such as being more volatile. That is, a researcher may not be able to determine when to use them correctly. The models do not tell the correlation within the model variables (Arora & Taylor, 2018). Additionally, the models are not declarative hence becoming more complex when analyzing data.
3. Smoothing constant
The smoothing constant usually enables the researcher to determine the weight of the historical time series values. Therefore, this constant will reduce or increase the past observed value and modify it to the future value in absolute magnitude.
Arora, S., & Taylor, J. W. (2018). Rule-based autoregressive moving average models for forecasting load on special days: A case study for France. European Journal of Operational Research, 266(1), 259-268.
Piepho, H. P. (2019). A coefficient of determination (R2) for generalized linear mixed models. Biometrical Journal, 61(4), 860-872.
student respond 2:
The coefficient of correlation is represented by the R-value in the summary table of a regression output.The value of this coefficient falls between -1 and +1. A negative correlation is represented by ” – ” and a postive correlation is represented by ” + “. To find the coefficient of determination you take the sqaure of the coefficient of correlation (R2) .The coefficient of determination helps us understand how the differences in one variable can be explained by a difference in another variable. It also gives the percentage of variation between two variables.
These two types of correlation are used in regression analysis to indentify the relationships between variables in data. Regression analysis is used to create a hypothesis to describe the relationship between a dependent variable and independent variable.
The key disadvantage to moving average forecasts models is that it is calculated using past data and doesn’t help with determing future data trends or changes. Another disadvantage is that in order to use moving average forecasts effectively, you must keep extensve records of pass data. In situations where this is not practical or easy, moving average forecast models will not be your best option. For example, a business that deals with massive amounts of data and must analyze them quickly might not benifit from using a moving average forecast model.
A smoothing forecast model is used to a create a time series. In this type of model, recent observations of data are given more weight and older observations of data decrease in weight. The value of the smoothing constant lies between 0 and 1. This value determines how much weight should be applied. For example, if the smoothing constant value is closest to 1, more weight will be placed on recent observation when compared to older observations. On the otherhand, if the value is closest to 0 than less wieght is placed on recent observations.This means when the smoothing cosntant value is closer to 0 there will be less difference between the weights of past and recent forecasts and observations.