Data Analytical and Forecasting Lecture 1

Business
Data
Analysis
&
Forecasting
Forecasting for Management
Decision
Dr. Asra Shaikh

Assistant Professor (Management
Sciences)
 Where do you see yourself in the next 5 years?
 Forecasting is the process of making statements about events
whose actual outcomes have not yet been observed.
 It involves predicting future outcomes of various business
decisions.
 Financial forecasts are fundamentally informed guesses, and
there are risks involved in relying on past data and methods
that cannot include certain variables.
 Helps the organization make plans that will lead to becoming a
financially successful business.
Forecasting is important for businesses as :
1.It helps organizations to make strategic plans for the
future.
2.Helps organizations to allocate resources
effectively and efficiently.
3.Plays a crucial role in financial planning by helping
businesses estimate future revenues, expenses, and profits.
4.Helps organizations to identify problems, potential risks,
and uncertainties and develop risk management strategies
to mitigate them.
5.Provides decision-makers with valuable insights and data,
which can help them make better-informed decisions and
develop short & long-term success strategies.
 Simple linear regression:
 Multiple linear regression:

1. SIMPLE LINEAR REGRESSION
2. MULTIPLE REGRESSION
3. MOVING AVERAGE
4. EXPONENTIAL SMOOTHING
5. CAUSAL MODEL
6. DELPHI METHOD
7. NAÏVE MODEL
8. ARTIFICIAL INTELLIGENCE OR NEURAL NETWORK
1. LINEAR REGRESSION (Compare 1 independent with 1
dependent variable)
2. MULTIPLE REGRESSION (Compare > 1 Independent with
1 dependent variable)
3. MOVING AVERAGE (takes averages of past data to
predict the future)
4. EXPONENTIAL SMOOTHING (weighted average of past
values)
5. CAUSAL MODEL (Cause and effect)
6. DELPHI METHOD (Opinion of experts)
7. NAÏVE MODEL (last period's actuals are used as this
period's forecast)
8. ARTIFICIAL INTELLIGENCE OR NEURAL NETWORK
(process data like the human brain)
Eg: Weather Forecasting: Analysing temperature, rainfall, and
other meteorological data
 Lag features are target values from previous periods.
 For example, if you would like to forecast the sales of a
retail outlet in period t you can use the sales of the
previous month t−1 as a feature. That would be a lag of 1
and you could say it models some
2017 kind 2018
of momentum.
 Lags are very useful in time series analysis because of a
phenomenon called autocorrelation, which is a tendency
for the values within a time series to be correlated with
previous copies of itself.
 One benefit to autocorrelation is that we can identify
patterns within the time series, which helps in determining
seasonality, the tendency for patterns to repeat at periodic
frequencies.
 lags and autocorrelation are central to numerous
forecasting models that incorporate autoregression,
regressing a time series using previous values of itself.
 Autoregression is the basis for one of the most widely used
forecasting techniques, the autoregressive integrated
moving average model or ARIMA for short.
 Objective of forecasting is to provide managers with
information that will facilitate decision making.
 Forecasting is an integral part of the planning and
control system, and organizations need a forecasting
procedure that allows them to predict the future
effectively (with certainty) and in a timely fashion.
 Top management is generally interested in making
decisions based on forecasting economic factors that
are critical in strategic planning and action.
 While forecasters will not be completely certain of
what will happen in the future, they can reduce the
range of uncertainty surrounding a business decision.
The art of forecasting is in recognizing when it is needed and how to
incorporate qualitative and quantitative data in the forecasting process.
The science of forecasting is embedded in the scientific principles of
model building. Forecasting often relies on statistical and mathematical
models that use historical data and patterns to make predictions. This is
the scientific aspect of forecasting, as it involves the use of data,
algorithms, and empirical evidence.
As in any scientific field, scientists begin with using the simplest
approach to explain a phenomenon.
 If the model is a good representation of the real-world conditions,
and its results do conform with observed phenomena, it is usually
accepted as an appropriate tool to predict the future.
 If, on the other hand, the simple model is not able to capture or
explain the observed phenomenon in detail, scientists use more
complex models.
 Forecasting is both an art and a science. It involves
a combination of quantitative analysis and
subjective judgment.
 In practice, the best forecasts often result from a
balance between the scientific and artistic aspects.
 Analysts use scientific methods to build a foundation
for their forecasts, but they also need the creativity
and judgment to adjust for unexpected events or to
consider qualitative factors that can't be easily
captured by data and formulas.
 Forecasting is a skill that involves both a structured,
data-driven approach and a certain degree of
intuition and insight.
 To develop accurate forecasts consistently, an
organization must utilize a systematic
procedure for forecasting that can be applied
quickly and modified as needed.
 The forecasting process can be either simple or

complex. It begins when management of an
organization requires an answer for its
management decision.
1. Download the data from world bank source regrading
the GDP, INFLATION, TRADE, all in (annual %)
2. LINK: https://data.worldbank.org/
3. Make separate sheets on excel of cross sectional,
timeseries and panel data sets (of any selected
countries)
4. Do the analyse the data as per line charts, pie charts or
bar charts (discussed in class).

Data Analytical and Forecasting Lecture 1

Uploaded by

Copyright:

Available Formats

Data Analytical and Forecasting Lecture 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Data Analytical and Forecasting Lecture 1

Uploaded by

Copyright:

Available Formats

Business

Dr. Asra Shaikh

 Multiple linear regression:

 The forecasting process can be either simple or

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.