Professional Documents
Culture Documents
Marketing Analytics PBL
Marketing Analytics PBL
ON
SHAIFALI AGRAWAL
SUBMITTED AT
DEPARTMENT OF BUSINESS
ADMINISTRATION
BUNDELKHAND INSTITUTE OF ENGINEERING
AND TECHNOLOGY JHANSI UTTAR PRADESH
1
DECLARATION
We undersigned, hereby declared that the project “A Study
On Modeling Trends And Seasonality” submitted in partial
fulfillment for the award of degree of master of business
administration of DR. A PJ ABDUL KALAM TECHNICAL
UNIVERSITY is a bonafide record of work done by me under
the guidance of (SHAIFALI AGRAWAL) (MBA Department)
this problem-based learning project report has not
previously formed the basis for the award of any degree,
diploma or similar title of any university.
Kaushambhi Tiwari(2100430700016)
Manish Dwivedi(2100430700021)
Md farhan(2100430700024)
Mohit kumar (2100430700025 )
2
CERTIFICATE
This is to certify that the problem-based learning project
report title “A Study On Modeling Trends And
Seasonality” being submitted by-
Kaushambhi Tiwari(2100430700016)
Manish Dwivedi(2100430700021)
Md farhan(2100430700024)
3
ACKNOWLEDGEMENT
We would like to express our special thanks of gratitude
to our professor S HAI FALI AGARWAL who gave us the
golden opportunity to do this wonderful project “.A
Study On Modeling Trends And Seasonality”.
4
INDEX
1. 6-11
MODELING TREND
2. 11- 14
SEASONALITY
3. 14- 18
4. 18
CONCLUSION
5. 19
BIBLIOGRAPHY
5
Modeling Trend
6
🔴Constant Patterns in Data: When looking at sales numbers, for
example, a constant trend is seen when there is no net increase or
decrease in sales over time. The sales may increase or decrease at
specific dates, but the overall average stays the same. However, even if
the average results are the same within a year, there still can be
seasonal changes. For example, sales levels may be consistently greater
in the summer and lower in the winter, although the average is the
same in the entire year.
7
🔴More Complicated Patterns: Trend forecasting can also deal with
patterns that are much more complicated than constant, linear and
exponential graphs. For example, a damped trend may show there was
an overall increase in sales for a number of years and then a sudden
stop. A polynomial trend might show a gradual increase, then
stagnation in sales over time followed by a decrease in sales.
8
LinkedIn is the largest professional networking service. Unlike
Facebook, LinkedIn is a business-oriented social network allowing its
members to place online resumes and to connect with other members
for career opportunities. Shows the number of LinkedIn members (in
millions) by quarter starting from the first quarter of 2009 and ending
with the second quarter of 2014.
The number of members are clearly increasing with time. However, the
trend appears not to be linear. Created using Excel’s trendline option
with charts, shows a linear trend fit to the LinkedIn series. It is clear
that the linear trend model fit is systematically off. The linear model is
not able to capture the curvature in the series. One possible approach
to fitting curved trends is to introduce the square of the time index to
the model:
ŷt=b0+b1t+b2t2
9
Su
ch a fitted model is known as a quadratic trend model. Shows the best-
fitting quadratic trend model. In comparison to the linear trend model,
the quadratic trend model fits the data series remarkably well.
10
Seasonality
Seasonality is a characteristic of a time series in which the data
experiences regular and predictable changes that recur every calendar
year. Any predictable fluctuation or pattern that recurs or repeats over
a one- year period is said to be seasonal.
11
Seasonality Types: There are three common seasonality types:
yearly, monthly and weekly.
12
🔴 Time-series are noisy. Random market fluctuations impact the sales,
and make the seasonality more difficult to isolate.
Seasonal patterns:
Variables of economic interest are often tied to other events that
repeat with regular frequency over time. Agriculture-related variables
will vary with the growing and harvesting seasons. Sales data may be
linked to events like regular changes in the weather, the start of the
school year, and the celebration of certain holidays. As a result, we find
a repeating pattern in the data series that relates to a particular
“season,” such as month of the year, day of the week, or hour of the
day. In the applications to follow, we see that to improve the accuracy
of our forecasts, we need to account for seasonal variation in the time
series.
13
Discussing time series forecasting, there are three types of time series
patterns: trend, seasonal, and cyclic. In this case, we just discuss the
trend and seasonal patterns. A trend pattern exists when there is a
long-term decrease or increase in the time series. The trend can be
linear or non-linear, such as an exponential function. In another hand,
seasonal pattern exists when the data is influenced by seasonal factors,
such as a day of the week, a month, or two-quarter of the year. In other
words, a seasonal pattern exists of a fixed known period.
Regression
Based on what we’ve discussed before, regression is one of the main
tasks in supervised machine learning. You need some input and your
target variable is a single floating-point number. For example,
predicting the world gold price. A big difference between regression
and classification is its targets. The targets in regression are just a few
infinite ones, while you have finite possible targets for classification.
14
Extra Trees Regression also known as Extremely Randomized Trees is a
meta estimator that fits several randomized decision trees on various
sub-samples of the data set and uses averaging to control over-fitting
and improve the predictive accuracy.
15
Seeing the two above modeling, we can decide to have two models:
One which predicts the trend (Figure 1) and another one which predicts
seasonality effects (Figure 2). Based on the two models above, can we
have a good “trend” and “seasonality” forecasting? The answer you can
find in the following section.
16
where f1(x) is Elastic Net Regression Function and f2(x) is Extra Trees
Regression Function.
2. Fit f1(x) on the training data set, where f1(x) has to predict the global
trend and thus solve the extrapolation problem.
y′ = y − f1(x)
17
global trend and local seasonality-effects. Below, you can see a list of
scoring functions by using MAE.
Con
clusion
MAE tells us how many the airline passengers prediction is typically
away from the true value. The small MAE value means good prediction
and vice versa. Please note that the high MAE value here do not mean
the regression model is bad. Some might be better when use more data
set, some might be better suited to different regression problems, e.g.
extrapolation instead of interpolation or working with higher
dimensional data. Thank you for reading this article and have a nice
weekend :)
BIBLIOGRAPHY :-
18
Modeling Trend https://theintactone.com/2020/01/18/modeling-
trend/
https://digfir-published.macmillanusa.com/psbe4e/
psbe4e_ch13_8.html
Seasonality https://theintactone.com/2020/01/18/seasonality/
https://medium.com/it-paragon/how-to-make-good-trend-and-
seasonality-forecasting-a163379d613f
19