SPSS Construction of the Company's Sales Trend : Report

SPSS CONSTRUCTION OF THE
COMPANY'S SALES TREND

TABLE OF CONTENTS
INTRODUCTION...........................................................................................................................1
Details of the area to be forecast..................................................................................................1
.....................................................................................................................................................1
Justification of the forecasting/modelling techniques which will be recommended...................2
Details of the data required to carry out the forecasts.................................................................2
Cost – benefit assessment of the options identified.....................................................................2
Gather appropriate data for the forecasting problem identified...................................................3
Findings.......................................................................................................................................3
Exponential smoothing approach of time series........................................................................11
CONCLUSION..............................................................................................................................14
Recomendation..............................................................................................................................15
REFERENCES..............................................................................................................................16
Figure 1Sales data............................................................................................................................1

INTRODUCTION
Trend analysis is the one of the most important approach that is used for making
prediction in the business. In the current report Gazpom company is taken which is well known
around the world for natural gas production and transportation. In the current report, different
approaches of time series are used for making predictions. It is observed that all approaches are
giving same results. Techniques applied for time series are ARIMA model, weighted moving
average and simple exponential smoothing method. There are some merits and demrits of these
approahces and same are taken in to consideration while doing time series analysis. At end of the
report, conclusion is formed and recommendation is made.
Details of the area to be forecast
Gazprom is one of Russian largest natural gas producer. It is involved in extraction,
production and transportation of natural gas (Open joint stock company gazprom 1999
consolidated financial reports, 1999). Demand of natural gas is highly affected by flcutuation in
its price and global economic conditions. Sales is prone to high level of uncertainity and due to
this reason sales is taken as area of forecast. In past years from 1999 to 2016
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
Chart Title
Figure 1Sales data
It can be seen from chart given above that sales increase from 1998 to 2007 constantly and then
it declined sharply for two years time period but recovery happened and from 2012 to 2013
moderate fluctuation happened in sales value. Second time again decline observed in sales which
1 | P a g e

was very moderate as sales tumbeled by moderate percentage and still there is very low growth
in sales. Hence, due to this reason sales value of the firm is considered for making prediction.
Justification of the forecasting/modelling techniques which will be recommended
There are three forecast modeling approaches that are recommended for making
prediction. ARIMA model: ARIMA model is recommended because it is an apprach under which
auto regression and moving average approaches are used for making predictions. It must
be noted that under this approach seasonal decomposition is also done and under this
different factors due to which fluctuation happened in data set are identified and removed
like seasonal and cyclical factors etc (Aoki, 2013). Hence, there is high reliability of
mentioned approach for making prediction. Weighted moving average approach: It is another apporach that is used to make
prediciton. Under moving average approach value of variable for specific time period is
taken in to consideration and average of same is taken. Hence, moving average approach
reflect extent to which variable fluctuate for certain time period (Box and et.al., 2015).
Under this approach to most recent observation heavy weight is given then to those who
are past observations in time series analysis. Thus, it can be said that weighted moving
average is best apporach then moving average method. Simple exponential smoothing: There are number of approaches that comes under
exponential smoothing and one of them is simple exponential smoothing. This method
can be applied on data set on which seasonal factor does not seems to be existed. In
current daat set also it is observed that there is absence of seasonality factor (O'Connor.
and et.al., 2010). Hence, it can be said that simple exponential smoothing also seem
appropriate for the firm.
Details of the data required to carry out the forecasts
In order to carry out forecast for the firm sales data was required. Relevant data is gathed
from year 1999 to 2017. This large size data is taken to ensure that appropriate number of factors
are taken in to account for making prediction. In order to ensure that data is analyzed in proper
manner relevant methods are used for making prediction like weighted average modeling of
prediction etc.
2 | P a g e