Ten weeks of data on the Commodity Futures Index are as follows.
1 answer:
The analysis of time series is a process by which a set of observations in a time series is analyzed.
Part a: The time series plot is constructed both in de$mo$ and excel sheet.
b. The data appear to follow a horizontal pattern.
Part b: The trial and error is used to find the value of exponential smoothing coefficient that results in MSE= 0.009 which is less than 0.01.
The MSE in the first table is calculated using the average of the two previous years.
The MSE in the second table is calculated using the moving average
The MSE in the third table is calculated by using ∝= 0.4 and the value of alpha changes to different values as shown in the cell U20, V20 and W20.
The MSE is the mean squared error.
We find the error by subtracting <u><em>forecast</em></u> data from the <em><u>actual</u></em> data.
The forecasted data may be average of 2 years , moving average of 3 years or adding the multiplied chosen value of alpha with the actual data and the multiplied value of 1- ∝ with the forecasted value of data.
The error is then squared.
The squared error is then averaged which gives the MSE.
The values of ∝= 0.400641005, 0.40319659, 0.400641001 may give the best MSE values.
<span class="sg-text sg-text--link sg-text--bold sg-text--link-disabled sg-text--blue-dark">
xlsx
</span>
You might be interested in
<u>Solution</u><u>:</u>
- First square both sides.
- Now, square root and square gets cancel out in the LHS. And in the RHS, apply the identity: (a + b)² = a² + 2ab + b².
- Now, transpose 4x and 4 to LHS.
- Now, do the addition and subtraction.
<u>Answer</u><u>:</u>
<u>x </u><u>=</u><u> </u><u>±</u><u> </u><u>3</u>
Hope you could understand.
If you have any query, feel free to ask.