Quantitative data analysis of the costs of direct programs

The costs of direct programs and their effects on the financial planning for upcoming months and the fiscal year in the United States will be the basis for this research paper’s quantitative data analysis. Placing realistic vote heads for each program will be aided by the availability of data on the running budget for required programs. The goal of the study is to establish a precise prediction of direct program costs for the fiscal year that ends on June 30, 2017. The essay will examine all operating expenses, including income, wages, travel expenses, client costs, conferences, and educational expenses (Matthew & Michael, 1986). Precisely, the operating costs for each program shall be examined on the basis of expenditure for every month from July 1st, 2016 to June 30th, 2017. In general, the quantitative approach in data analysis has proven invaluable in gratifying the main objective of this study. Through QA, we have successfully forecasted the cost of direct programs in a business entity.

Keywords: Direct Program Cost, MAD, MSE, MAPE, Moving Average, and Exponential Smoothing.

Direct Program Costs

As guides Matthew & Michael, (1986), all the steps of the analysis were followed by first defining the problem in question and subsequently giving the primary objective of the study. Based on the research we have done, there is always extra expenses that cannot be covered even in miscellaneous budgets. These costs can put companies at a risk of bankruptcy or overdraw the accounts before the budgeted period is over. The main causes of unplanned expenditure include any form of calamities, the war on cyber-attacks especially on the point of sale systems, and other current trends in financial management threats.

Since data has been availed representing direct program cost in a sample home group in the United States, this study’s objective is to forecast the cost of direct programs in the year 2017 (Matthew & Michael, 1986). This cost is inclusive of all revenues, salaries, client salaries, conference and education cost, occupancy costs, other office expenses, total insurance cost, vehicle expenses, travel costs, and administrative expenses. An in-depth exploration of these costs shall be set to distinguish the different types extra costs associated with each vote head. This in-depth study shall aim to provide a suitable forecast of the required adjustments towards each vote in order to accommodate any unplanned expenses.

Development of the model used in data collection and the actual attainment of the raw data was done. Guided by our objectives, the forecasting model adopted in this study was meant to give out the best predictions. For all the raw data examined, we ran the proficiency of obtaining moving averages. The follow-up was to implement weighted moving averages model to assign various weights to the previously collected data and the observations made from that data, as understood from Matthew & Michael (1986). It was key to run the weighted moving average because it can assist in laying a credible emphasis on the present cost trends. The final step was the use of exponential smoothing. The model combines both the previous forecast and the last observed value.

During the analysis process, each of the three models shall be studied to examine the most viable one. The Mean Square Error (MSE) and Mean Absolute Deviation (MAD) are the statistical measures to be studied in this case. In performing MAD, our primary focus shall be to determine how accurate the financial forecast is while in implementing MSE, we shall compute the average values of the squared error terms of the forecasted model. The combination of these two statistical measures shall give a foundation for determining the accuracy of the forecast (Matthew & Michael, 1986).

Both weighted moving averages and moving averages were both ran for the two models discussed above. The weights we applied were one and two for the weighted moving average model with 2 used for the recent period. Additionally, exponential smoothing model had alpha applied as the smoothing constant. Two dummy variables, 0 and 1 were used to represent two seasons; winter and summer seasons. With 0 representing summer and 1 corresponding to the winter season. Furthermore, α, the smoothing constant is set to be 0.5 to all the models used in smoothing. If alpha is greater than 0.5, a skewed distribution is produced. It is, therefore, necessary to use 0.5 so that an even distribution is attained.

The table below shows a case study of direct costs in a financial management program. The actual total cost is 340, 066.30 compared to the prior YTD 338, 119.17. The preceding year costs figures have also been placed to cover the uncertainties likely to happen as seen from the deviation from prior YTD to the actual YTD.

Table 1. Direct program costs

Month

Cost figures in $

Estimate cost in $

Deviation

July

24, 746.92

24, 586.72

-160.20

August

25, 987.22

25, 824.96

-162.26

September

27, 334.83

26, 149.48

-1,182.35

October

26, 440.43

27, 174.83

743.40

November

26, 044.48

27, 014.83

970.35

December

31, 755.36

27, 177.09

-4,578.27

January

31, 727.24

31, 566.24

-161.00

February

28, 532.20

31,728.50

3,034.04

March

27, 150.17

28, 695.30

1,545.13

April

29, 011.26

26, 987.91

-2,023.45

May

29, 472.98

29, 173.52

-299.46

June

31, 863.21

29, 333.52

-2,529.69

The data provided herein is according to The Institute of Professional Practice, Inc.

1-01-0015 Davis for the fiscal year ending June 30th, 2017. The projected costs alongside all actual costs running direct programs have been provided. The deviations from the estimated cost of the actual costs of direct program costs have a high degree of inaccuracy in predicting the future costs direct programs. For this study, the models used are going to provide an extremely accurate forecast to prevent the kind of sharp deviations evident from the table 1. above. From that table, the calculated exponential smoothing is $ 27, 626.03, the weighted moving average is $1798.91, while the moving average is $ 28, 383.95. These figures are estimated for the whole financial year beginning July 1st, 2016 to June 30th, 2017. The involved standard error was 0.5. These calculated values are a forecast for the in 2016-2017 representing the rate at that fiscal year.

In order to implement the models chosen in this study, the beginning point was to analyse the direct costs of production about the total revenues collected. The data below represents the values of MAD, MAPE, and MSE as weighted against the said variables. The data presented below shows all the calculations made on the forecasting model. From this table 2, the values of MAD, MSE, and MAPE have been determined.

Table 2: The data on forecasting models

MONTH

ACTUAL COST (A)

FORECAST(F)

ERROR (A-F)

|A-F|

(A -F)^2

|(A-F)÷ A|

Jul-16

24746.92

24586.72

-160.2

160.2

25664.04

0.00647353

Aug-16

25,987

25824.96

-162.26

162.26

26328.3076

0.00624384

Sep-16

27334.83

26149.48

-1,182.35

1182.35

1397951.52

0.04325434

Oct-16

26440.43

27174.83

743.4

743.4

552643.56

0.02811603

Nov-16

26044.48

27014.83

970.35

970.35

941579.123

0.03725742

Dec-16

31755.36

27177.09

-4,578.27

4578.27

20960556.2

0.14417314

Jan-17

31727.24

31566.24

-161

161

25921

0.0050745

Feb-17

28532.2

31728.5

3,034.04

3034.04

9205398.72

0.1063374

Mar-17

27150.17

28695.3

1,545.13

1545.13

2387426.72

0.05691051

Apr-17

29011.26

26987.91

-2,023.45

2023.45

4094349.9

0.06974706

May-17

29472.98

29173.52

-299.46

299.46

89676.2916

0.01016049

Jun-17

31863.21

29333.52

-2,529.69

2529.69

6399331.5

0.07939219

MAD was calculate using the formula Σ (|A - F| ÷ n). This formula gives the value MAD as $ 1449.133 for the data used in the forecast process. The formula Σ ((A – F) 2) ÷ n was used in calculating MSE, which gave a value of 3,842,235.57 obtained from the average of squared errors. Finally, MAPE was obtained by applying the formula; Σ ((| (A-F) ÷ A|) ÷ n) × 100, from which the value of MAPE was gotten as 0.04942837 × 100 = 4.9428%.

All the analyses above clearly indicate that the weighted moving average of $1798.91 per fiscal year is the most accurate model for all direct program costs in the financial year ending June 30th, 2017. This forecast shows that the projected cost of direct programs is expected to rise to an MSE of 1,280,745.19 per the fiscal year in the forthcoming years. The analysis of the data in table 2 above gave us the opportunity to conclude that the cost of a direct product, as points out Risk Management Manual (1997).

Since there are diverse forms of businesses, this study gives an open end for other researchers to use other models for forecasting on the direct program costs. An absolute way this research could have adopted was the quantitative analysis that uses mathematical models in real-world known as simulation. This study is ideal for producing the direction most companies should take for them to control financial uncertainties. The mathematical simulation model can accommodate a wider range of variables and could be shed some more light on forecasting the expenditure of direct programs.

References

A Guide to the Project Management Body of Knowledge (PMBOK) Third Edition, (2009). An American National Standard, ANSI/PMI. Project Management Institute Inc.

Bureau of Labour Statistics, (2012). U.S. Department of Labour, Occupational Outlook Handbook, 2012-13 Edition, Cost Estimators, on the internet at https://www.bls.gov/ooh/business-and-financial/cost-estimators.html.

CIPS. (2016). Procurement and supply workflow, Published by Profex Publishing Ltd., p. 49

GAO Cost Estimating and Assessment Guide, Best Practices for Developing and Managing Capital Program Costs, (2009). United States Government Accountability Office.

Matthew B.M. & Michael, H. (1986). Quantitative Data Analysis: A Sourcebook of New Methods. Educational Evaluation and Policy Analysis, 8 (3); 329-331.

Risk Management Manual. (1997). International Federation of Consulting Engineers, p. 52.

Standard Estimating Practice Sixth Edition, (2004). American Society of Professional Estimators, Bni Publications, Inc.