Construction Management Report

ConstructionManagement Report

InstitutionAffiliation

Parta)

Examiningthe operating earnings using regression model

Hypothesistesting

At95% t critical = ±1.96

NullHypothesis: ß = 0

AlternativeHypothesis: ß ≠ 0

OUTPUT

Multiple R

0.931

R Square

0.867

Adjusted R Square

0.826

Standard Error

38.126

Number of observations

60

ANOVA

d.o.f

SS

MS

F

Significance F

Regression

14

426750.607

30482.186

20.97

0.000

Residue

45

65411.117

1453.580

Total

59

492161.724

Coefficients

Standard Error (S.E)

t- start = Coef/S.E

P – value

Significance (-1.96 ≤ t ≤ 1.96

Intercept

-364.267

98.345

-3.704

0.001

Significance

Size

0.771

0.099

7.803

0.000

Significant

EMPL

-0.866

1.466

-0591

0.558

Not significant

Total

-0.010

0.013

-0.805

0.425

Not significant

P15

0.057

0.028

2.050

0.046

Significant

P25

0.013

0.013

0.978

0.333

Not significant

P35

0.014

0.022

0.658

0.514

Not significant

P45

0.001

0.033

0.032

0.975

Not significant

P55

0.010

0.014

0.727

0.471

Not significant

INC

8.763

1.644

5.331

0.000

Significant

COMP

-2.681

2.320

-1.156

0.254

Not significant

NCOMP

-0.347

1.561

0.222

0.825

Not significant

NREST

1.451

0.245

5.920

0.000

Significant

PRICE

-3.173

0.966

-3.283

0.002

Significant

CLI

0.402

0.700

0.574

0.569

Not significant

AdjustedModel

Thismodel has been obtained by eliminating the insignificant independentvariables. These are the variables that do not have a significanteffect on the operating earnings of the stores. The aim of modifyingthis previous model is to minimize ambiguity of the model byexcluding irrelevant variables in the model. The adjusted model,therefore, looks like

OUTPUT

Multiple R

0.909

R Square

0.826

Standard Error

38.126

Number of observations

60

ANOVA

d.o.f

SS

MS

F

Regression

14

426750.607

30482.186

20.97

Residue

45

65411.117

1453.580

Total

59

492161.724

Coefficients

Standard Error (S.E)

t- start = Coef/S.E

P – value

Intercept

-364.267

98.345

-3.704

0.001

Size

0.771

0.099

7.803

0.000

P15

0.057

0.028

2.050

0.046

INC

8.763

1.644

5.331

0.000

NREST

1.451

0.245

5.920

0.000

PRICE

-3.173

0.966

-3.283

0.002

Partb)

Evaluatingthe stores that I would have opened in 1994.

Asa transition manager, I would have considered the target performanceration set by the company to assess the viability of each store. Thiscan be gotten from the formula

Performanceratio &gt0.26

store

EARN

K

Ratio

51

216.3

776

0.279

52

65.7

648

0.101

53

67.6

690

0.098

54

127.9

715

0.179

55

82.9

650

0.128

56

-2.9

788

-0.004

57

247.7

782

0.317

58

343.0

1558

0.220

59

193.1

936

0.206

60

277.5

688

0.403

Therefore,the above analysis has revealed that it is only three store that hadmet the company’s target performance ratio of 0.26 = 26%. Thestores that could have been opened in the year 1994 are store number51, store number 57 and store number 60.

Partc

Selectingthe best locations for the stores

Ihave applied regression model in site selection. It is important toconsider resource utilization that selecting the best locations forthe stores. In this case, I have suggested three stores to be locatedat three different location. The best sites that I have choseninclude Toulouse, Marseilles-2, and Montpellier. The areas selectedare based on the regression model, and therefore I have consideredthe significance of each independent variables since they have asignificant impact on the profitability of the stores,

Partd)

Memorandum

TO:TO THE MANAGEMENT CROQPAIN COMPANY

FROM:MICHAEL (TRANSITION MANAGER)

SUBJECT:LOCATION SELECTION FOR THE STORES

Themultiple regression model is an analysis that helps to describe therelationship between several independent and a dependent variable.Regression model assumes that the relationship between variables islinear. This model offers a comparative criteria, and this is why Ihave suggested to use the model in selecting locations stores wherethe store can be opened. Regression model helps in predicting thefuture. The primary strength of the regression model is inforecasting what is likely to happen in future. By analysing thetrend or the past performance, I was able to easily predict theexpected performance of the store in various locations.

Anotherpower of the model is that it helps in supporting decisions. Sinceselection is made based on the past trend and the company’stargets, the decision is, therefore, supported by the analysis of thefindings. For example, through regression model, I was in a positionto determine the most appropriate locations where new stores can beconstructed. The model also helps to summarize the large amount ofdata to a simplified one that can be easily understood by the usersat all levels. Regression model helps in correcting errors. Despitefrom supporting the manager’s decision, the model also helps inadjusting the way of thinking. For example, one might think that thecost of living index has a significant influence on the operatingearnings of the business but through regression model, we have seenthat this variable is insignificant in determining the earnings.

However,there are certain difficulties when dealing with multiple regressionsthat pose a dilemma for the comparative research. The comparativemodel may mismatch the variable and hence result in misinterpretationof the model. The model also assumes that the relationship betweenthe variables is linear though in some cases this may not be thecase.

Despitethe weaknesses mentioned above, regression model remains the bestcriterion to use in selecting locations for construction of the storeas compared to other models such as least square method. Therefore,as a transition manager, I would recommend the application of thismodel.

Yoursfaithful,

M.M

Michael.

References

Construction_Department_at_CrogPain.pdf

LindA., Marchal G. &amp Wathen A. (2011). StatisticalTechniques in business and Economics.New Yolk: McGraw Hill