Naïve Method, Average of Historic Cost and Moving Average Method

You might also like

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 2

Naïve Method, Average of Historic Cost and Moving

average method
The most important factor of forecasting is analysing which model to use and which one
is the best fit for the data or organisation.

To compare the lowest of MFE MAE MSE MAPE is observed.

To compare the fits of different time series models. Smaller values indicate a better fit. If
a single model does not have the lowest values for all 3 accuracy measures, MAPE is
usually the preferred measurement.

  MFE MAE MSE MAPE


Naïve
Method 0.68 9.83 172.87 32.93%
Avg of
Historical 5.69 9.31 141.90 524.96%
Moving
Average -0.20 7.37 139.80 30.19%

Here, MFE is positive for the Naïve and Avg. of historic data method, implying that the
method is generally under forecasting, here observed value is greater than forecasted
value.
But in case of Moving average, it is negative, which implies the observed value is
lower than forecasted leading to over forecasting.
MAE is the real absolute values of the error and hence moving average is better than
the two other methods.
As we can’t find one method with the lowest error values, we will tend to go with
MAPE as it is the most reliable source, hence Moving average method should be
preferred by the organisation.

Exponential Soothing

Exponential smoothing models use smoothing weights to determine the amount of


emphasis to give to each data point when estimating the final values of the model
components

The weights can take on any values between 0 and 1

Small values (e.g., between 0.0 and 0.1) indicate that the given component is changing slowly
and a lot of smoothing is being applied when estimating its final value. Large values (e.g.,
between 0.85 and 1.0) indicate that the given component is changing rapidly and a lot of
emphasis is being given to the recent data when estimating its final value to capture the
rapid change in the data

The forecasts should follow the general flow of the data at the end of the series. Because
the forecasts from single exponential smoothing are constant, it is important that there
is no trend in the data before the forecasts. If there is a trend before the forecasts, the
forecasts may not be accurate.

We have set alpha to 0.4, the previous data point is given a relatively small weight
while the previous smoothed value is given a large weight (i.e. 0.9).

The graph shows an increasing trend. Excel cannot calculate the smoothed value for
the first data point because there is no previous data point.
The smoothed value of the second data point equals the previous data point.
With this forecast firm can make plans and decisions accordingly,

Since MSE factor plays an important role in calculating the error, the lowest MSE i.e.,
0.4 alpha is chosen with 129.73 alpha.

Least square Method


Least squares regression is used to predict the behaviour of dependent
variables.

the Least Squares Method can help to quantify the relationship between two
or more variables: such as a Net sale and the forecasted sales. It is very
helpful in forecasting as it reduces the error .

You might also like