Forecasting.gap

1

Alpha Construction Group

Housing Components Products Demand ForecastNew Developments in Measurement and Analytics

2

Forward

• Many businesses are affected by the cyclical effects of

housing and remodeling demand, yet they have not

developed formal tools to allow these data to drive their

business plans and forecasts.

• Global-Analytics-Partners has significant experience in

this and can provide clients with an interactive tool which

will help them more accurately forecast and plan their

business performance.

3

Content

• Forecast Model Description

• Sales Predictions v. Actuals

• Forecast Summary and Assumptions for 2015

• Drivers of Growth for 2014

• Model Driver Relative Importance

• Model Driver Impacts and Sensitivities– Recurring Seasonality

– Housing Starts

– Remodeling Spending

– Order Backlog

– Price to Customer

• Forecast-Simulator Demonstration

• Summary

4

• The forecast model covers monthly periods from Jan 2009 to present. Four positive

drivers and one negative driver were used.

• The positive drivers include recurring seasonality, housing starts, Home Remodeling

Spend & Order Backlog. The negative driver was average price to the customer.

Home

Remodeling

Spend

Order

Backlog

Price-to-

Customer

Monthly

Unit

Demand

Housing

Starts

Recurring

Seasonality

5

-

20,00,000

40,00,000

60,00,000

80,00,000

100,00,000

120,00,000

Jan

.2006

May.2

006

Se

p.2

006

Jan

.2007

May.2

007

Sep

.2007

Jan

.2008

May.2

008

Se

p.2

008

Jan

.2009

May.2

009

Se

p.2

009

Jan

.2010

May.2

010

Sep

.2010

Jan

.2011

May.2

011

Se

p.2

011

Jan

.2012

May.2

012

Se

p.2

012

Jan

.2013

May.2

013

Sep

.2013

Jan

.2014

May.2

014

Se

p.2

014

Jan

.2015

May.2

015

Se

p.2

015

Forecast/Simulated Sales

Actual Sales

Forecast & Actual Sales Plot

Model R2=93.2%, Mean Absoluter Error +/- 4.1%

From the bottom of the recession in 2009, Components have been growing at a +9.5% annual rate per year.

6

Forecast Summary & Assumptions

• With modest but slightly improving growth in key drivers,

plus flat pricing, we would expect 2015 components

demand to improve to about +7.2% in 2015.

• This would generate near-term record components sales

of over 116 MM.

7

Sources of Growth

-1.0%

-0.4%

-0.04%

2.0%

2.2%

-1.5% -1.0% -0.5% 0.0% 0.5% 1.0% 1.5% 2.0% 2.5%

Price to Cust

Drawings

Repairs

Housing Starts

Baseline Momentum

Growth Impact

Growth Impact

• Based on the model, here is the contribution to Housing Components overall annual growth of +2.8%

8Relative Impact Importance of Housing

Component Business Drivers

33%

6%

25%

1%

35%

Importance of Sales Impact

Recurring Seasonality

Price to Customer

Housing Starts

Order Backlogs

Remodel Spend

Remodel Spend, followed by Seasonality and Housing Starts comprise the dominant

drivers of components’ demand.

9

Impact of Recurring Seasonality

8784

88

99 101

109 111114 113 111

98

83

40,00,000

45,00,000

50,00,000

55,00,000

60,00,000

65,00,000

70,00,000

75,00,000

80,00,000

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Average Sales Impact of Recurring Seasonality

The “high season” for component demand is June-October

10

Price-to-Customer Demand Sensitivity

86,00,000

87,00,000

88,00,000

89,00,000

90,00,000

91,00,000

92,00,000

93,00,000

94,00,000

-6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0%

Mo

nth

ly D

em

an

d

Change in Price

Monthly Demand Impact from Change in Price to Customer

• The sensitivity of price-to-customer is slightly “inelastic” at -0.6, meaning a 1% increase would yield about a -0.6% decline in sales.

• This sensitivity indicates that price increases are likely to be modestly profitable. On average there is also about a 3 month lag between the largest change in sales and the change in price.

11

Housing-Starts Demand Sensitivity

50,00,000

60,00,000

70,00,000

80,00,000

90,00,000

100,00,000

110,00,000

-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%

Mo

nth

ly D

em

an

d

Change in Housing Starts

Monthly Demand Impact from Change in Housing Starts

• Housing starts can vary significantly from period to period. Overall, every 10% increase in starts generates about a +2.8% increase in Components sales.

• On average there is also about a 3 month lag between the largest change in sales and the change in housing-starts.

12

Remodeling Spend Demand Sensitivity

50,00,000

60,00,000

70,00,000

80,00,000

90,00,000

100,00,000

110,00,000

-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%

Mo

nth

ly D

em

an

d

Change in Drawings

Monthly Demand Impact from Change in Remodel Spend

• Remodeling Spending by Home Owners has the largest single impact of any driver. Every 10% change is connected to a 4.6% increase in component sales.

• On average there is also about a 3 month lag between the largest change in sales and the change in remodeling spending

13

Monthly Order Backlog Demand Sensitivity

90,00,000

90,10,000

90,20,000

90,30,000

90,40,000

90,50,000

90,60,000

90,70,000

90,80,000

90,90,000

91,00,000

-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%

Mo

nth

ly D

em

an

d

Change in Repairs

Monthly Demand Impact from Change in Order-Backlogs

• Backlogs have the least impact of any driver.

• There are also no lead-lag relationships between demand and

changes in repairs

14

Forecast-Simulation Tool

• Below is a rendering of the interactive-forecasting tool

which will be demonstrated

15

Summary

• Across 5 key business drivers, we have derived a model which connects to monthly Components demand with a 93% fit.

• For 2014, prior sales momentum and housing starts are the two leading positive drivers of annual +3.1% sales gain.

• With “modest” assumptions about the growth in key drivers, and flat pricing, for 2015, we expect to see demand growth of about +7.2% versus about +3.1% growth in 2014.

• Drawings, recurring seasonality and housing starts form the most dominant drivers of components monthly demand

16

Bangalore, IN Office:No. 141, 2nd Cross, 2nd Main,Domlur, 2nd Stage, Bangalore 560071Phone: +91 80 40917572, +91 80 [email protected]

Contact Us US Office:

Suite 100, 1780 Chadds Lake Dr, NE

Marietta, Georgia, 30068-1608

Atlanta, USA

[email protected]