Question

Part 1: From the choice of simple moving average, weighted moving average, exponential smoothing, and linear regression analysis, which forecasting technique would you consider the most accurate? Define each method. Why did you choose this one?

Answer 1

Answer: Find the explanation below.

Step-by-step explanation:

1a. Simple Moving Average: is derived by summing the last data set (example, prices), and dividing it by the number of periods (n) expected for the mean. For example, for a data set containing last prices in the order; 10, 20, 30, 50, we simply get the sum of the prices and divide by the number of prices we have. This would mean 110/4 = 27.5

1b. Weighted Moving Average: is obtained by first weighting the given data set. The prices are then multiplied by the weighted values and then, summed up. The sum of the weighted values must equal 1.

1c. Exponential Smoothing: is a forecasting method that calculates the weighted average of past observations with more emphasis on the most recent data set.

1d. Linear Regression Analysis derives the linear relationship between a dependent variable and one or more independent variables. An equation like

Z = a + b*p explains this relationship.

Where Z = dependent variable

a = constant

b = regression coefficient

p = independent variable.

2. I would choose Exponential Smoothing as the best way to forecast data because it deals with more recent data, and as such, it can more accurately predict fluctuations in prices. The Linear regression analysis depends on predictors so validating the relationship between predictors and the data being worked on might be difficult.

Answer 2

Answer: Weighted Moving Average

Step-by-step explanation:

The Simple Moving Average as a Moving Average is one of the core indicators in technical analysis.

It is the easiest one of the Moving Averages and is simply calculated by taking the AVERAGE price over a specified period.

It is called a "moving" average because it is plotted on the chart bar by bar, forming a line that moves along the chart as the average value changes.

The Weighted Moving Average is like the Simple one but instead it assigns heavier weighting to more current data points because they are seen as more relevant. The sum of these weights should add up to 1. It is essentially not equally distributed like a Simple Moving Average. It is calculated by multiplying the given price by its associated weighting and totaling the values.

Exponential Smoothing is one of numerous window functions that can be applied to smoothen data in signal processing and as such is used as a rule of thumb technique for smoothing time series data using the exponential window function.

Finally, Linear Regression Analysis is a very common type of predictive analysis and so is very widely used.

When using Linear Regression, the basic questions we are asking ourselves is;

a) do a set of independent variable/s do a good job of predicting an outcome / dependent variable?

b)which variables are the most significant in predicting the outcome variable and in what way do they impact the outcome of the variable. This is done through the use of the sign and magnitude of beta estimates.

Of these four choices, the Weighted Moving Average is the MOST ACCURATE.

Why?

It is because using this method, one can place weights on variables in accordance with their importance thereby enabling a better measure of variables.

There is a drawback though, the Weighted Moving Average method can be cumbersome to use if a longer time span is taken because the user would have to keep adjusting the weights.

It is still the Most Accurate however because the other measures make Assumptions such as an Averages, Variables need to be in a Straight line, or Exponential curve but the Weighted Average may be modified to any form.

If you need any clarification do react or comment.