MODULE:_7__

MODULE GALS

Familiarize with different

concepts of Diagnostic
and Predictive Analytics BUSINESS ANALYTICS:

Describe performance and PREDICTIV

E
operations of businesses
using diagnostic and PREPARED BY: MR. DAN JEWARD C. RUBIS, MBA
predictive analytics

ANALYTICS
PREDICTIVE
ANALYTICS
 Each type of data analysis can help you reach specific goals and be used in tandem
to create a full picture of data that informs your organization’s strategy formulation
and decision-making.
WHAT IS
PREDICTIVE
ANALYTICS
PREDICTIVE
ANALYTICS
is the use of data to predict future trends and events. It uses historical data to
forecast potential scenarios that can help drive strategic decisions.

The predictions could be for the near future—for instance, predicting the
malfunction of a piece of machinery later that day—or the more distant future,
such as predicting your company’s cash flows for the upcoming year.
PREDICTIVE
ANALYTICS
Predictive analysis can be conducted manually or using machine-learning
algorithms. Either way, historical data is used to make assumptions about the
future.

Forecasting can enable you to make better decisions and formulate data-
informed strategies.
WHAT IS
FORECASTING?
Forecasting
is a technique that uses historical data (demand , sales, budget and etc.) as inputs to make
informed estimates that are predictive in determining the direction of future trends.
WHAT ARE THE COMMON FEATURES TO ALL
FORECASTS?
FORECASTING
1. Forecasting technique generally assume that the same underlying causal system that
existed in the past will exist in the future.

2. Forecasts are rarely perfect; predicted values usually differ from the actual results.

3. Forecasts for group of items tend to be more accurate than forecast for individual
items.

4. Forecast accuracy decreases as the time period covered by the forecast increases.
 FORECASTING is the art and science
of predicting what will happen in the
future. Sometimes that is determined by
a mathematical method; sometimes it is
based on the intuition of the operations
manager. Most forecasts and end
decisions are a combination of both.
FORECAST DECISIONS
Forecasts affect decisions and activities throughout an organization

Accounting Cost / Profit estimates

Finance Cash flow and funding

Human Resources Hiring / Recruiting / Training

Marketing Pricing, Promotion, Strategy

MIS IT / IS systems, services

Operations Schedules, MRP, workloads

Product / Service Design New products and services

TYPES OF
FORECASTING
FORECASTING
ECONOMIC FORECASTS
address the business cycle.
They predict housing starts,
inflation rates, money
supplies, and other
indicators.
FORECASTING
TECHNOLOGICAL FORECASTS
monitor rates of technological
progress. This keeps organizations
abreast of trends and can result in
exciting new products. New
products may require new facilities
and equipment, which must be
planned for in the appropriate time
frame.
FORECASTING
DEMAND FORECASTS
deal with the company's products
and estimate consumer demand.
These are also referred to as sales
forecasts, which have multiple
purposes. In addition to driving
scheduling, production, and
capacity, they are also inputs to
financial, personnel, and marketing
future plans.
CATEGORIES
FORECASTING
QUALITATIVE
Qualitative forecasting techniques
are subjective, based on the opinion
and judgment of consumers and
experts; they are appropriate when
past data are not available. They
are usually applied to intermediate-
or long-range decisions.
EXECUTIVE JUDGEMENT (TOP DOWN)

Groups of high-level executives will often assume responsibility

for the forecast. They will collaborate to examine market data
and look at future trends for the business. Often, they will use
models as well as market experts to arrive at a forecast
SALES FORCE OPINIONS (BOTTOM UP)

The sales force in a business are those persons most close to

the customers. Their opinions are of high value. Often the sales
force personnel are asked to give their future projections for
their area or territory. Once all of those are reviewed, they may
be combined to form an overall forecast for district or region.
DELPHI METHOD

A group of experts are recruited to participate in a forecast.

The administrator of the forecast will send out a series of
questionnaires and ask for inputs and justifications. These
responses will be collated and sent out again to allow
respondents to evaluate and adjust their answers
MARKET SURVEYS

Some organizations will employ market research firms to solicit

information from consumers regarding opinions on products
and future purchasing plans.
QUANTITATIVE
Quantitative forecasting models are
used to forecast future data as a
function of past data. They are
appropriate to use when past
numerical data is available and when
it is reasonable to assume that some
of the patterns in the data are
expected to continue into the future.
Naive Forecast
a forecast that uses the actual sales for the past period as forecasted sales for the next period.
Naive Forecast
a forecast that uses the actual sales for the past period as forecasted sales for the next period.
Example: Use Naive method to determine the forecast sales.

Period (2019) Sales Forecast

Jan 65k
Feb 68k

Mar 65k

April 64k

May 68k

June 70k

What is the naïve forecast for the month of July?

Moving Average
uses the most recent n data values in the time series forecast for the next period.
Formula: Moving Average = ∑(most recent n data values) ÷
n Illustrative example 1:
Period Actual Demand Forecast
Jan 210
Feb 250
Mar 290
April 210
May 250
June 210
July 180

Calculate for a 2 -month moving average using the data in the given table.
What is the 2-month moving average forecast for the month of August?
Moving Average
Illustrative example 2:

Period Actual Demand Forecast

Jan 210
Feb 250
Mar 290
April 210
May 250
June 210
July 180

Calculate for a 3 - month moving average forecast using the data in the given table.
What is the 3 - month moving average forecast for the month of August?
Time-Series Forecasting – Moving Average
F = MA4

YEAR SALES (ACTUAL) SALES ERROR

In Millions (FORECAST) (A-F)
In Millions

2005 5 -
2006 4 -
2007 4 -
2008 7 -
2009

What is the 4-year moving average forecast for the year 2009?
Time-Series Forecasting – Moving Average
F = MA4

YEAR SALES SALES ERRO

(ACTUAL) (FORECAST) R
In Millions In Millions
(A-F)
2005 5 -
2006 4 -
2007 4 F = (5+4+4- +7) / 4

2008 7 -
2009 8 5 3
2010
Time-Series Forecasting – Moving Average
F = MA4

YEAR SALES (ACTUAL) SALES ERROR

In Millions (FORECAST) (A-F)
In Millions

2005 5 -
2006 4 -
2007 4 -
2008 7 -
2009 8 5 3
2010

What is the 4-year moving average forecast for the year 2010?
Time-Series Forecasting – Moving Average
F = MA4

YEAR SALES SALES ERROR

(ACTUAL) (FORECAST) (A-F)
In Millions In Millions

2005 5 -
2006 4 -
2007 4 -
2008 7 F = ( 4+4+7+8) / 4
-
2009 8 5 3
2010 5.75

What is the 4-year moving average forecast for the year 2010?
Time-Series Forecasting – Moving Average
F = MA5

YEAR SALES (ACTUAL) SALES ERROR

In Millions (FORECAST) (A-F)
In Millions

2003 4 -
2004 6 -
2005 5 -
2006 8 -
2007 9 -
2008

What is the 5-year moving average forecast for the year 2008?
Time-Series Forecasting – Moving Average
F = MA5

YEAR SALES (ACTUAL) SALES ERROR

In Millions (FORECAST) (A-F)
In Millions

2003 4 -
2004 6 -
2005 5 -
2006 8 -
2007 9 -
2008 5 6.4 (1.4)
Time-Series Forecasting – Moving Average
F = MA5

YEAR SALES (ACTUAL) SALES ERROR

In Millions (FORECAST) (A-F)
In Millions

2003 4 -
2004 6 -
2005 5 -
2006 8 -
2007 9 -
2008 5 6.4 (1.4)
2009

What is the 5-year moving average forecast for the year 2009?
Time-Series Forecasting – Moving Average
F = MA5

YEAR SALES SALES ERRO

(ACTUAL) (FORECAST) R
In Millions In Millions
(A-F)
2003 4 -
-
2004 6 F = (4+6+5+ 8+9) / 5

2005 5 -
2006 8 F = (6-+5+8+9+5) / 5

2007 9 -
2008 5 6.4 (1.4)
2009 4 6.6 (2.6)
WEIGHTED MOVING AVERAGE
 WEIGHTED MOVING AVERAGE

Use the following weights: Week Sales Weighting Weighted Average

0.40 = Most recent (A =
0.35 = 2nd most recent Actual)
0.25 = Oldest date 1 18
1.00 TOTAL
2 13

3 16

SUM
WEIGHTED MOVING AVERAGE

Use the following weights: Week Sales Weighting Weighted Average

0.40 = Most recent (A =
0.35 = 2nd most recent Actual)
0.25 = Oldest date 1 18 0.25
1.00 TOTAL
2 13 0.35

3 16 0.40

SUM 1
 WEIGHTED MOVING AVERAGE

Use the following weights:

Week Sales Weighting Weighted Average
0.40 = Most recent
(A =
0.35 = 2nd most recent Actual)
0.25 = Oldest date 1 18 0.25 4.5
1.00 TOTAL
2 13 0.35 4.55

3 16 0.40 6.40

SUM 1 15.45
WEIGHTED MOVING AVERAGE

Use the following weights: Week Sales Weighting Weighted Average

0.40 = Most recent (A =
0.35 = 2nd most recent Actual)
0.25 = Oldest date 1 11
1.00 TOTAL
2 15

3 13

SUM
 WEIGHTED MOVING AVERAGE

Use the following weights:

Week Sales Weighting Weighted Average
0.40 = Most recent
(A =
0.35 = 2nd most recent Actual)
0.25 = Oldest date 1 11 0.25 2.75
1.00 TOTAL
2 15 0.35 5.25

3 13 0.40 5.2

SUM 1 13.20
WEIGHTED MOVING AVERAGE
Assume that we want to calculate the weighted moving average of five stock prices over a 5-
day period. The prices are P58.91, P61.52, P59.32, P55.43, and P54.65, with the last price
being the most recent.

Day Price Weighting Weighted Average

1 58.91

2 61.52

3 59.32

4 55.43

5 54.65

SUM
WEIGHTED MOVING AVERAGE
Assume that we want to calculate the weighted moving average of five stock prices over a 5-
day period. The prices are P58.91, P61.52, P59.32, P55.43, and P54.65, with the last price
being the most recent.

Day Sales Weighting Weighted Average

1 58.91 1/15

2 61.52 2/15

3 59.32 3/15

4 55.43 4/15

5 54.65 5/15

SUM 1
WEIGHTED MOVING AVERAGE
Assume that we want to calculate the weighted moving average of five stock prices over a 5-
day period. The prices are P58.91, P61.52, P59.32, P55.43, and P54.65, with the last price
being the most recent.

Day Sales Weighting Weighted Average

1 58.91 1/15 3.93

2 61.52 2/15 8.20

3 59.32 3/15 11.86

4 55.43 4/15 14.78

5 54.65 5/15 18.22

SUM 1 56.99
 Weighted Moving Average
is a smoothing method that uses a weighted average of the recent n data as the forecast.
Formula: WMA = ∑ (weight for period n)(demand in period n) ÷ ∑ weights
Month Demand Forecast
Jan 21
Feb 25
March 29
April 21
May 25
June 20
July 18
August 21
September 20

1. Compute for a 3 - month weighted moving average forecast of the given table.
2. What is the 3- month weighted moving average forecast for the month of October?
 Weighted Moving Average
is a smoothing method that uses a weighted average of the recent n data as the forecast.
Formula: WMA = ∑ (weight for period n)(demand in period n) ÷ ∑ weights
Month Demand Forecast
Jan 21
Feb 25
March 29
April 21
May 25
June 20
July 18
August 21
September 20
1. Complete the third column of given table.
2. What is the 4- month weighted moving average forecast for the month of October?
Exponential Smoothing
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

ES = 0.40 (Actual (Forecast) (A-F)
)
1 65
2 55
3
4
5

F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)

PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN:
ES = 0.40 F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Week

1
2
3
4
5
Exponential Smoothing
GIVEN:
ES = 0.40 F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK

Week

1
2
3
4
5

SAMPLE WEEK 3
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
= (0.40)(55) + (1-0.40)(65)
= (0.40)(55) + (0.60)(65)
= 22 + 39
= 61
Exponential Smoothing
GIVEN: Week Sales Sales Forecast Error
(Actual) (Forecast) (A-F)
ES = 0.40
1 65
2 55 65 (10)
3 61
4
5

F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)

PREVIOUS WEEK PREVIOUS WEEK

SAMPLE WEEK 3
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
= (0.40)(55) + (1-0.40)(65)
= (0.40)(55) + (0.60)(65)
= 22 + 39
= 61
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

ES = 0.40 (Actual (Forecast) (A-F)
)
1 65
2 55 65 (10)
3 58 61
4
5

F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)

PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

(Actual (Forecast) (A-F)
ES = 0.40
)
1 65
2 55 65 (10)
3 58 61 (3)
4
5
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

(Actual (Forecast) (A-F)
ES = 0.40
)
1 65
2 55 65 (10)
3 58 61 (3)
4 59.8
5
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

(Actual (Forecast) (A-F)
ES = 0.40
)
1 65
2 55 65 (10)
3 58 61 (3)
4 64 59.8 4.2
5
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

(Actual (Forecast) (A-F)
ES = 0.40
)
1 65
2 55 65 (10)
3 58 61 (3)
4 64 59.8 4.2
5 61.48
F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

ES = 0.30 (Actual (Forecast) (A-F)
)
1 65 60 5
2 55 ? ?
3 58 ? ?

F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)

PREVIOUS WEEK PREVIOUS WEEK
Exponential Smoothing

GIVEN:
ES = 0.30 F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)
PREVIOUS WEEK PREVIOUS WEEK
Week

1
2
3
Exponential Smoothing

GIVEN: Week Sales Sales Forecast Error

(Actual (Forecast) (A-F)
ES = 0.30
)
1 65 60 5
2 55 61.50 (6.5)
3 58 59.55 (1.55)

F = (ES)(ACTUAL SALES) + (1 – ES)(SALES FORECAST)

PREVIOUS WEEK PREVIOUS WEEK
EXPONENTIAL SMOOTHING
is a time series forecasting method for univariate data that can be extended to support data
with a systematic trend or seasonal component.

Univariate
is a term commonly used in statistics to describe a type of data which consists of observations
with only one characteristic or attribute.
A simple example of univariate data:
a. Number of workers in BPO industry.
b. Demand for a particular product.
c. Company’s Budget for next year.
d. Sales of a company.
Exponential smoothing formula:
Forecast for the current period = Forecast in the Last Period + α (Actual value of demand or
sales – Forecast in the last period)
Ft = Ft - 1 + α [ A t -1 – Ft - 1 ]
Where:
Ft = new forecast
Ft - 1 = forecast of the demand, sales, customers and etc of the last period.
.: α = smoothing constant has a value between 0 to 1
A t - 1 = actual demand, sales, customers, and etc. of the last period.
Note: If you set the value of α = 0, basically your forecast for next period is equal to the
forecast of the last period. Using the appropriate value of α is the key to have an accurate
value in forecasting
How are we going to
determine the value of
alpha in exponential
smoothing?
Rules:
1. If the actual demand, sales, number of customers and etc, of the last period appears to be relatively
stable over a period of time then select a relatively smaller value of alpha (α) which closer to zero (0)

Demand

Time
2. If the demand, sales, number of customer and etc, tends to fluctuate rapidly then select a relatively
larger value of alpha (α) closer to one (1).

Demand
Time
Illustrative example 1:
Illustrative example 1:
In March 2021 a demand for 200 units of Toyota car model “Wigo” for April has forecasted by a car dealer. Actual April
demand was 250 cars. Forecast the demand for May using exponential smoothing constant of α = 0.30 and α = 0.70
Given:
Ft - 1 = 200 cars previous forecast
α = 0.30 and α = 0.70
A t - 1 = 250 cars previous period’s actual demand
Solution: α = 0.70 α = 0.30
Ft = Ft - 1 + α [ A t -1 – Ft - 1 ] Ft = Ft - 1 + α [ A t -1 – Ft - 1 ]
Ft = 200 + (0.70) [ 250 – 200 ] Ft = 200 + (0.30) [ 250 – 200 ]
Ft = 200 + (0.70) [ 50 ] Ft = 200 + (0.30) [ 50 ]
Ft = 200 + 35 Ft = 200 + 15
Ft = 235 cars Ft = 215 cars
Illustrative example 2:
Illustrative example 2:
In January 2021 a demand for 2,000 units of computers for February has predicted by computer store owner. Actual
February 2021 demand was 3,500 units . Forecast the March demand using an exponential smoothing constant of
α = 0.80 and α = 0.20
Given:
Ft - 1 = 2,000 units of computer (previous forecast)
α = 0.80 and α = 0.20
A t - 1 = 3,500 units of computer (previous period’s actual demand)
Solution: α = 0.80 α = 0.20
Ft = Ft - 1 + α [ A t -1 – Ft - 1 ] Ft = Ft - 1 + α [ A t -1 – Ft - 1 ]
Ft = 2,000 + (0.80) [ 3,500 – 2,000 ] Ft = 2,000 + (0.20) [ 3,500 – 2,000 ]
Ft = 2,000 + (0.80) [ 1,500 ] Ft = 2,000 + (0.20) [ 1,500 ]
Ft = 2,000 + 1,200 Ft = 2,000 + 300
Ft = 3,200 units of computer Ft = 2,300 units of computer
 SAMPLE PROBLEM
You are the Finance Manager of XY Corporation. Given the following data, you were asked to calculate
forecasts for months 4, 5, 6, and 7using a three-month moving average and an exponential smoothing
forecast with an alpha of 0.3. Assume a forecast of 61 for month 3:

FORECAST FORECAST
MONT ACTUAL
(MOVING AVERAGE) (EXPONENTIAL
H SALES
SMOOTHING)
1 56
2 61
3 58 61
4 67
5 75
6 76
7
THANK YOU!