CHAPTER

Forecasting
Introduction
• Forecasting is the art and science of predicting
the future
• A forecast is a starting point and a base for any
kind of planning and decision making process
• The objective of forecasting is to reduce risk in
decision making
• The survival of a business enterprise depends
on its ability to assess the market/demand
trend with reasonable accuracy a head time
 What is Forecasting?
 Forecasts are predictions, projections or
estimates of future events or conditions in the
environment in which an enterprise operates.
Or
 Forecasts are estimates of occurrence, timing
and magnitude of uncertain future event
 What is Forecasting?
 Process of predicting
uncertain future event
 Underlying basis of ??
all business decisions
 Production
 Inventory
 Personnel
 Facilities
 What is Forecasting?
 The estimates can be
 Occurrence of an event
 Occurrence time
 Magnitude or volume
The purpose of forecasting is to use the best
available information to guide future activities
toward organizational goals.
 What is Forecasting?
 Forecasting can be used to determine
 Demand of product or service
 prices or costs
 Change of technology
 Competitors
 Interest rate
 Exchange rate
 Good forecasts enables managers to plan and
budget for appropriate levels of personnel, raw
materials, capitals, inventory and a lot of other
variables.
 Forecasting Time Horizons
 Short-range forecast
 Up to 3 months
 Job scheduling, job assignments, purchasing,
workforce levels, production levels
 Medium-range forecast
 3 months to 3 years
 Sales and production planning, budgeting
 Long-range forecast
 3+ years
 New product planning, facility location,
research and development
 Distinguishing Differences
Medium/long range forecasts deal with more
comprehensive issues and support management
decisions regarding planning and products,
plants and processes
Short-term forecasting usually employs different
methodologies than longer-term forecasting
Short-term forecasts tend to be more accurate
than longer-term forecasts
 Influence of Product Life Cycle
 Product life cycle is the length of time from a product
first being introduced to consumers until it is
removed from the market
 A product's life cycle is usually broken down into four
stages: Introduction – Growth – Maturity – Decline
 As product passes through its life cycle, forecasts are
useful in projecting
 Staffing levels
 Inventory levels
 Factory capacity
 Product Life Cycle
Introduction Growth Maturity Decline
Product design and Forecasting critical Standardization Little product
development Product and process Less rapid product differentiation
critical reliability changes – more Cost
Frequent product minor changes minimization
OM Strategy/Issues

Competitive product
and process design improvements and Optimum capacity Overcapacity in
changes options the industry
Increasing stability
Short production Increase capacity of process Prune line to
runs eliminate items
Shift toward product Long production
High production focus runs not returning
costs good margin
Enhance distribution Product
Limited models improvement and Reduce capacity
Attention to quality cost cutting
 Product Life Cycle
Introduction Growth Maturity Decline
Best period to increase Practical to change Poor time to change Cost control critical
market share price or quality image image, price, or quality
Company Strategy/Issues

R&D engineering is Strengthen niche Competitive costs

critical become critical
Defend market position
CD-ROMs
Analog TVs
Drive-through
restaurants

Sales iPods

3 1/2”
Xbox 360 Floppy
disks
 Types of Forecasts
 Economic forecasts
 Address business cycle – inflation rate,
money supply, etc.
 Technological forecasts
 Predict rate of technological progress
 Impacts development of new products
 Demand forecasts
 Predict sales of existing products and
services
 Strategic Importance of
Forecasting

 Human Resources – Hiring, training, laying

off workers
 Capacity – Capacity shortages can result in
undependable delivery, loss of customers,
loss of market share
 Supply Chain Management – Good supplier
relations and price advantages
Commonly Observed Normal Demand
patterns/trends
Constant
Pattern in which data values
fluctuate around a constant mean. Linear Trend
Pattern in which data
exhibit increasing or
decreasing values
over time.

Any pattern that

regularly repeats
itself and is Data patterns created by
Seasonal Cyclical economic fluctuations
constant in length
Abnormal Demand Patterns
Random demand variation is unexplained variation
that cannot be predicted

Sudden
Transient Sudden
Fall
Impulse Rise
 Forecasting Methods

Qualitative Methods Quantitative Methods

Delphi Method
Time series causal
Jury of Executive Average
Opinion
Linear Regression
Naïve Method

Sales Force Multiple Regression

Moving Average
Composite

Weighted Moving
Average
Analogies
Exponential
Market Survey smoothing

Single ES

Double ES
 Selection of Forecasting Method

• Some techniques of forecasting are quite simple and

rather inexpensive to develop and use, whereas others
are extremely complex, require significant amounts of
time to develop, and may be quite expensive
• Some are best suited for short-term projections,
whereas others are better for intermediate-term, or
long-term forecasts
 ……..cont'd
What technique or techniques to select depends on the
following criteria
1. Cost of developing forecasting model
2. How complicated are the relationships that are
being forecasted?
3. Is it for short-run or long-run purposes?
4. How much accuracy is desired
5. Minimum tolerance level of errors
6. How much data is available
 Qualitative Methods
• Qualitative forecasting methods, are methods in which
the forecast is made subjectively by the forecaster
• They are educated guesses by forecasters or experts
based on intuition, knowledge, and experience
Useful:
 There is no enough time to gather and analyze
quantitative data
 When political and economic conditions are changing,
available data may be obsolete and more up-to-date
information might not yet be available
 The introduction of new products and the redesign of
existing products
 ….…Cont'd
There are different ways of forecasting the future
demand based on qualitative methods
1. Delphi Method
2. Jury of Executive Opinion
3. Market Survey
4. Opinions of Salesperson
5. Historical Analogy and Life Cycle Analysis
 ……Cont'd
1. The Delphi Method
• Delphi method is a forecasting method in which the
objective is to reach a consensus among a group of
experts while maintaining their anonymity
a. A researcher/coordinator puts together a panel of
experts in the chosen field. These experts do not
have to be in the same facility or even in the same
country. They do not know who the other panelists
are.
 ……Cont'd
b. The coordinator prepares questionnaires and
sends it to each expert on the panel
c. Each expert makes brief statements of predictions
d. The coordinator collects the written predictions
together and summarize the findings
e. Then The coordinator prepares an updated
questionnaire incorporating the findings and
sends it to each expert
• This process continues until a consensus is
reached
 ……Cont'd
• The Delphi method has the advantage of not
allowing anyone to dominate the consensus
• It takes a large amount of time
• It is found an appropriate method for forecasting
long-range product demand, technological change,
and scientific advances in medicine
 ……Cont'd
2. Jury of Executive Opinion
•The subjective views of executives or experts from
sales, production, finance, purchasing and etc. are
averaged to generate a forecast about future sales
•This method is often used for strategic forecasting or
forecasting the success of a new product or service.
•This method is fast, less expensive
Disadvantages
•The forecast bases subjective opinions
•Often the opinion of one person can dominate the
forecast if that person has more power than the other
members of the group or is very domineering
 ……Cont'd
3. Market Survey
• Is an approach that uses surveys and interviews to
determine customer likes, dislikes, and preferences and
to identify new-product ideas
• The method involves the use of questionnaires,
interviews, formation of consumer panels, and testing
of new products and services
 The purpose of the survey is to identify the nature of
consumer consumption
 A forecast is developed after determining how general
sales vary with differences in market location, buyer
occupation, commodity prices, quantity, quality,
consumer income and other factors
 ……Cont'd
4. Opinions of Salesperson
 This approach involves the opinions of the sales force
and these opinions are primarily taken into
consideration for forecasting future sales
 The sales people, being closer to the consumers, can
estimate future sales in their own territories more
accurately
 These forecasts are good for short range planning since
sales people are not sufficiently sophisticated to predict
long-term trends
 ……Cont'd
5. Historical Analogy and Life Cycle Analysis
 Market survey can be supplemented by reference to
the performance of an ancestor of the product or
service by applying product life cycle analysis
 Most products pass through stages of introduction,
growth, maturity and decline
Quantitative Methods
 These techniques use statistical analysis and other
mathematical models to predict future events,
primarily based upon past activities and historical
data.
 This method can be divided into two categories:
i. Time Series Forecasting Methods
ii. Causal/Associative Forecasting Methods
 ……Cont'd
i. Time Series Forecasting Methods
• Time series observations taken at regular intervals
• The intervals could be hourly, daily, weekly, monthly,
quarterly, or annually
• Forecasting techniques based on time-series data are
made on the assumption that past activities/time series
observations are good indication of the future
ii. Causal/Associative Forecasting Methods
• Rely on identification of related variables that can be
used to predict values of the variable of interest
• Development of an equation/mathematical model that
summarizes the effects of predictor/independent
 ……Cont'd
• Causal models can be very complex, especially
if they consider relationships among many
variables
• It requires model building
• Time series models are generally easier to use
and can generate a forecast more quickly than
causal models
 ……Cont'd
1. Naïve Method
– The naïve method is the simplest forecasting model
– It assumes that the next period's forecast is equal to
the current period's actual
Ft +1 = At
Where: Ft +1 = forecast for next period, t + 1
At = actual value for current period, t
t = current time period
 ……Cont'd
• It works well when there is little demand
variation from one period to the next
• The naïve method can be modified to take a
trend into account “naïve method with trend”
• One advantage of the naïve method is that it
is very simple
• Disadvantage
It ignores past historical data and relies on most
recent data
……Cont'd
……Cont'd
 Simple Moving Average Problem (1)

D
t
-1+D
t
-2+D
t-
3 +
...
+Dt
-n
F
t=
n
W eek D em and
1 650 Question: What are the
2 678
3 720
3-week and 6-week
4 785 moving average
5 859 forecasts for demand?
6 920
7 850 Assume you only have 3
8 758 weeks and 6 weeks of
9 892
10 920
actual demand data
11 789 for the respective
12 844 forecasts
 Calculating the moving averages gives us:

Week Demand 3-Week 6-Week

1 650 F4=(650+678+720)/3
2 678 =682.67
3 720 F7=(650+678+720
4 785 682.67 +785+859+920)/6
5 859 727.67
=768.67
6 920 788.00
7 850 854.67 768.67
8 758 876.33 802.00
9 892 842.67 815.33
10 920 833.33 844.00
11 789 856.67 866.50
12 844 867.00 854.83
 Simple Moving Average Problem (2)

W eek D em and Question: What is

1 820 the 3 and 5 week
2 775 moving average
3 680
forecast for this
4 655
data?
5 620
6 600 Assume you only
7 575 have 3 weeks and
5 weeks of actual
demand data for
the respective
forecasts
Simple Moving Average Problem (2) Solution

Week Demand 3-Week 5-Week

1 820 F4=(820+775+680)/3
2 775 =758.33
3 680 F6=(820+775+680
4 655 758.33 +655+620)/5
5 620 703.33 =710.00
6 600 651.67 710.00
7 575 625.00 666.00
 ……Cont'd
4. Weighted Moving Average
Forecasting method in which n of the most recent
observations are averaged by giving different weights
It assigns different weighting factor to old demands

F
=
t D
w
1
t+
-
1 D
w
2
t+
-
2wD
3
t+
-
3...
+
w
D
nt
-
n
Where: Ft = Forecast of demand for next period, t
Dt = Actual value/demand in period, t
Wt = weighting factor
n

w
i=1
i =1
 Weighted Moving Average Problem (1) Data

Question: Given the weekly demand and weights,

what is the forecast for the 4th period or Week 4?

Week Demand Weights:

1 650 t-1 .5
2 678 t-2 .3
3 720
t-3 .2
4

Note that the weights place more emphasis on

the most recent data, that is time period “t-1”
Weighted Moving Average Problem (1) Solution

Week Demand Forecast

1 650
2 678
3 720
4 693.4

F4 = 0.5(720)+0.3(678)+0.2(650)=693.4
 Weighted Moving Average Problem (2) Data

Question: Given the weekly demand information

and weights, what is the weighted moving
average forecast of the 5th period or week?

Weights:
Week Demand
t-1 .7
1 820
2 775 t-2 .2
3 680
4 655 t-3 .1
Weighted Moving Average Problem (2) Solution

Week Demand Forecast

1 820
2 775
3 680
4 655
5 672

F5 = (0.1)(755)+(0.2)(680)+(0.7)(655)= 672
Exponential smoothing (ES)/ Single ES
 In the previous methods of forecasting the major
drawback is the need to continually carry a large
amount of historical data
 Exponential smoothing is the most used of all
forecasting techniques. It is well accepted for the
following major reasons
 Exponential models are relatively accurate
 Formulating an exponential model is relatively easy
 The user can understand how the model works
 Little computation is required to use the model
 …Cont'd
 In this method, only three pieces of data are needed
to forecast the future: the most recent forecast, the
actual demand that occurred for that forecast period
and smoothing constant alpha(α)
 The smoothing constant determines the level of
smoothing and the speed of reaction
 The selection of smoothing constant depends upon
the characteristics of demand. High values of α are
more liable to fluctuations in demand. Low values of
α are appropriate for relatively stable demand
 …Cont'd
 The equation for a single exponential smoothing
forecast is simply

t
F D

t1

1F

t
1 
Where:
F t = The exponentially smoothed forecast for period t
Ft 1 = The exponentially smoothed forecast made for
the prior period t-1
D t 1 = The actual demand in the prior period t-1
 = The desired response rate or smoothing constant
0α1
 Example
Determine forecast for
periods 7 & 8
Period Actual
2-period moving average
1 300
4-period moving average 2 315
2-period weighted moving 3 290
average with t-1 weighted 4 345
0.6 and t-2 weighted 0.4 5 320
6 360
Exponential smoothing
with alpha=0.2 and the 7 375
period 6 forecast being 375 8
 Solution
2- 4- 2- Expon.
Period Actual Period Period Per.Wgted. Smooth.

1 300

2 315

3 290

4 345

5 320

6 360

7 375 340.0 328.8 344.0 372.0

8 367.5 350.0 369.0 372.6

Example
A furniture manufacturing company
produces different sizes of cabinets and
records the demand monthly. The following
demand data are for a specific cabinet
model: January 60; February 55; March 70.
Using 56 as the forecast for January and a
smoothing constant of 0.20, what is the April
sale? Is 0.20 a good choice as a smoothing
constant?
 Solution
F
tD

t1

1
F

t1
F 
Mar0
.
2

55
1
0.
205
56
.6
F
0
.
2
60
Feb 

1
0.
2056 57
56
.8
F
0.
2
Apr
70

1
0.
205
 57 59
.6
 60
The April forecast is 60 units (fractional units should be rounded
to be realistic). And the given smoothing constant (0.20), recent
demand is not weighted heavily. This seems appropriate for this
data. If demand is unstable, a higher constant value should be
used.
Trend-Adjusted Exponential Smoothing/
Double ES
• Uses three equations. The first smoothes out the
level of the series, the second smoothes out the
trend, and the third generates a forecast by adding up
the findings from the first two equations
– Step 1 – Smoothing the level of the series

F
tt
αD 
t1
α)(F
(1 T
t)
1
– Step 2 – Smoothing the trend/Trend estimate
T
tt
β(F
F

t1
) 
β)T
(1
t1

– Step 3 – Forecast including the trend

F1
t t
F T
t
 …Cont'd
F t = The Exponentially smoothed forecast for period t
T t = The Exponentially smoothed trend for period t
Ft 1 = The forecast including trend for period t+1
Ft 1 = The forecast including trend for period t-1
Dt = The actual demand of the prior period
 = smoothing coefficient of the average
 = smoothing coefficient of the trend
 Example
A company uses exponential smoothing with
trend to forecast usage of its lawn care products.
At the end of July the company wishes to
forecast sales for August. July demand was 62.
The trend through June has been 15 additional
gallons of product sold per month. Average sales
have been 57 gallons per month. The company
uses alpha+0.2 and beta +0.10. Forecast for
August. Make a forecast including trend for the
month of August
 Solution
Smooth the level of the series:

F
αD
(1
Jul
t y
t
1
α)(F
T)

t
1





0.2
62

0.8
57
15
70

Smooth the trend:


TF
Jul
tt
β(F
)

1
(1
y
t
1
β)T




0.1
70

57

0.9
15
14

Forecast including trend:

F
F
T 
70
14.8
August
t t 84.8
gall
 Example
Assume an initial starting Ft of 100 units, a
trend of 10 units, an alpha of 0.2 and a beta
of 0.3. If actual demand turned out to be 115
rather than the forecast 100, calculate the
forecast for the next period
Causal Forecasting Methods
• Casual/Associative techniques rely on
identification of related variables that can be
used to predict values of the variable of interest
• Investigating/exploring the relationships between
the variable of interest and related variables
• Develop a mathematical model, which is used to
forecast future values of the variable of interest
• Regression analysis is a statistical technique of
modeling the relationship between variables
 …Cont'd
 This method of forecasting has a higher degree of
accuracy due to the fact that it measures the
causal factors and asserts their relationships to
the products under consideration
 It is used for short and medium range forecasting
of existing products, services, marketing
strategies, production and facility planning.
 The cost of this method is relatively medium
 Regression Analysis
• Regression analysis is a statistical technique of modeling
the relationship between variables
There are two types of regression model:
Simple linear regression and Multiple regression
• Simple linear regression, is a regression model that
contains only one regressor/predictor/independent variable
- It is used to model the relationship b/n two variables
• Multiple regression is a regression model that contains
more than one regressor/predictor/independent variable
- It is used to model the relationship among three and
more variables
 Simple Linear Regression Model
The simple linear regression model takes the form of the
following straight-line relationship:


- Regression Equation
Y 01 X- Regression Line
- Regression Analysis equation
Where : - Mean value function model
Y = the dependent variable
X = independent variable
 0 and  1 = the intercept and the slope of
the line respectively (regression coefficients)
The mean/expected value of the dependent variable, Y is
assumed to be a linear function of the independent variable, X
 …Cont'd
• The actual observed value y may not fall exactly
on the straight line.
• The actual value of y is determined by the mean
value function (the linear model) plus a random
error term, 

Y= 0 + 1 X + 
Non random Random error
Errors in Regression
Y
the  
observed
data
point
ˆ
Y 
0
1X the
fitted
Y
i
. regress
lin

Yi

ii
Error
Y
{
ˆ
Y
i
t
Y
ihep
r
edi
cte
dva
l
ueo
fYf
orX
i

X
Xi
 …Cont'd
• Estimation of a simple linear regression relationship
involves finding Regression coefficients, 0 and 1
of the linear regression line
• The estimates of 0 and 1 should result in a line
that is (in some sense) a “best fit” to the data
• The values of Regression coefficients (0 and 1 )
should be determined so that the sum of the
squares of the vertical deviations is minimized.
• This criterion for estimating the regression
coefficients is called method of least squares.
The least squares method
 …Cont'd
 Notationally, it is occasionally convenient to give
special symbols to the numerator and denominator of
the previous Equation

  
Therefore;  1 = Sxy/Sxx ˆ
Y 0
1 X
Miles
1211
Dollars
1802
Miles 2
1466521
Miles*Dollars
2182222
Example
1345 2405 1809025 3234725
1422 2005 2022084 2851110
1687 2511 2845969 4236057
S xx   x 
2  x 2
1849 2332 3418801 4311868 n
2026 2305 4104676 4669930
79,448 2
2133 3016 4549689 6433128  293,426,946   40,947,557 .84
2253 3385 5076009 7626405 25
2400 3090 5760000 7416000  x ( y)
S xy   xy   
2468 3694 6091024 9116792 n
 390,185,014  (79,448)(106,605)  51,402,852 .4
2699 3371 7284601 9098329
2806 3998 7873636 11218388
25
3082 3555 9498724 10956510
S
3209 4692 10297681 15056628
b  XY  51,402,852 .4  1.255333776  1.26
3466 4244 12013156 14709704 1 S 40,947,557 .84
3643 5298 13271449 19300614
XX
 
3852 4801 14837904 18493452
b  y  b x  106,605  (1.255333776 ) 79,448 
4033 5147 16265089 20757852 0 1 25  25 
4267 5738 18207288 24484046
 274 .85
4498 6420 20232004 28877160
4533 6059 20548088 27465448
4804 6426 23078416 30870504
5090 6321 25908100 32173890
5233 7026 27384288 36767056
5439 6964 29582720 37877196
79,448 106,605 293,426,946 390,185,014
 Correlation coefficient (r)
• Correlation coefficient (r) measures the direction
and strength of the linear relationship between
two variables
• Correlation can range from -1 to +1
• The correlation between two variables can be
computed using the following equation

r

n 
XY X
Y
n 
 * n Y
  
2 2
X X 2
Y 2
 …Cont'd
• The square of correlation coefficient, r2 provides
a measure of the percentage of variability in the
values of dependent variable y that is
“explained” by the independent variable x
• The possible values of r 2 range from 0 to 1
• A relatively high value of r 2, would indicate that
the independent variable is a good predictor of
values of the dependent variable.
• A low value of r 2, would indicate a poor
predictor
 Example
A maker of golf shirts has been tracking the relationship
between sales and advertising dollars. Use linear
regression to find out what sales might be if the
company invested $53,000 in advertising next year
Sales $ Adv.$ XY X^2 Y^2
(Y)
130
(X)
32 4160 2304 16,900 b
XYnXY
1
X nX 2 2

2 151 52 7852 2704 22,801

3 150 50 7500 2500 22,500

28202 447.25147.25

b  1.15
9253 447.25
2
4 158 55 8690 3025 24964

153.85 53 a  Y  bX  147.25 1.1547.25

5
a  92.9
Tot 589 189 28202 9253 87165
Y  a  bX  92.9 1.15X
Avg 147.25 47.25 Y  92.9 1.1553  153.85
 …Cont'd
.
n XY
XY
r

nX  
X * nY Y 
2 2
2 2

428,202
189
589

r .982
* 487,165
589

2 2
4(9253)
-(189)
r .982
 .964
2 2
 Forecast Accuracy
• One of the most important criteria for choosing a
forecasting model is its accuracy
• Forecast error is the difference between the
forecast and actual value for a given period,
Et = At - Ft
Where: Et = forecast error for period t
At = actual value for period t
Ft = forecast for period t
• Note that over-forecasts = negative errors and
under-forecasts = positive errors
 …Cont'd
• There are a variety of possible sources of forecast
errors, including the following:
1. The model may be inadequate due to
 The omission of an important variable
 A change or shift in the variable that the model cannot
deal with (e.g., sudden appearance of a trend or cycle)
 The appearance of a new variable (e.g., new
competitor).
2. Irregular variations may occur due to severe weather or
other natural phenomena, temporary shortages or
breakdowns, catastrophes, or similar events.
 …Cont'd
3. Random variations. Randomness is the inherent
variation that remains in the data after all causes of
variation have been accounted for. There are always
random variations.
• A one time period forecast error provides a forecast
error of a particular, single period which fails to
asses forecasting performance of a given forecasting
technique
• The model's accuracy can be assessed only if
forecast performance is measured over time
 Measure of forecast accuracy
1. Mean Absolute Deviation


MAD

actual
fore
(MAD)- is the average of the sum n
of the absolute errors

2. Mean Square Error (MSE)-


actual
- 
2
forec
Measure of forecast error that 
MSE
computes error as the average of n
the squared error.
CFE
3. Tracking Signal-Measures if TS
your model is working MAD
4. Cumulative Forecast Error
(CFE)-Measures any bias in the CFE  
actu
for 
forecast
Example
The Instant Paper Clip Office Supply Company sells and delivers
office supplies to companies, schools, and agencies within a 50-
mile radius of its warehouse. The office supply business is
competitive, and the ability to deliver orders promptly is a big
factor in getting new customers and maintaining old ones.
(Offices typically order not when they run low on supplies, but
when they completely run out. As a result, they need their orders
immediately.) The manager of the company wants to be certain
that enough drivers and vehicles are available to deliver orders
promptly and that they have adequate inventory in stock.
Therefore, the manager wants to be able to forecast the demand
for deliveries during the next month. From the records of previous
orders, management has accumulated the following data for the
past 10 months:
 …Cont'd
a) Compute the monthly demand forecast for April through
November using a 3-month moving average.
b) Compute the monthly demand forecast for June through
November using a 5-month moving average.
c) Compute the monthly demand forecast for April through
November using a 3-month weighted moving average. Use
weights of 0.5, 0.33, and 0.17, with the heavier weights on the
more recent months.
d) Compute the mean absolute deviation for June through
October for each of the methods used. Which method would
you use to forecast demand for November?

Month Jan Feb Mar Apr May June July Aug Sep Oct
Orders 120 90 100 75 110 50 75 130 110 90
Solution
…Cont'd
Example
A company is comparing the accuracy of
two forecasting methods. Forecasts using
both methods are shown below along with
the actual values for January through May.
The company also uses a tracking signal
with ±4 limits to decide when a forecast
should be reviewed. Which forecasting
method is best?
 Solution
Method A Method B
Month Actual F'cast Error Cum. Tracking F'cast Error Cum. Tracking
sales Signal Error Signal
Error

Jan. 30 28 2 2 2 27 3 3 1

Feb. 26 25 1 3 3 25 1 4 1.5
Marc 32 32 0 3 3 29 3 7 3
h
April 29 30 -1 2 2 27 2 9 4
May 31 30 1 3 3 29 2 11 5

MAD 1 2
MSE 1.4 4.4