Forecasting and business analysis

An important graduate quality that is stressed in this course is the ability to conform to acceptable community and business standards. Such standards include most notably the completion of

professional-quality work on time. Excuses such as last-minute printer or file storage problems will not be accepted, as they would not be by a client in the real world. Students with extreme

medical or personal emergencies may seek an extension on an assignment through your tutor. Students who wish to seek an extension on the basis of an ongoing medical or learning disability

must present their Disability Access Plan to the course coordinator at least two weeks before the assignment is due, and preferably as soon as possible after the start of the study period. This

will enable us to best tailor an assessment plan to fit your circumstances.

Assignment 2 – ECON 2007
Word limit: 1000 (Acceptable to write up to 1200). Graphs and tables do not count into the number of words. Due Date: 13th June, 5PM Internal students should submit two copies of assignment:

One soft copy using Learnonline, one hard copy to either my pigeon hole (next to WL 2-21C) or the School of Commerce office (WL 2-57). Both must be received before the due date. External

students are required to submit the soft copy only. There is a penalty of 10% of the total mark for late submission of each day. Assignments submitted on the due date but after 5PM are

considered as one day late.

Executive Summary/Abstract: Briefly describe what your report is about. Summarise the results. Introduction: Introduce the topic. Data, Model & Method: -Data Description: What are RGRT, AWE

and UR? Where do you get all the data? How many observations do you have?… – Data must not be in different frequencies ? convert UR into quarterly data by taking the AVERAGE across the

three months in a quarter, and convert RGRT by taking the SUMMATION. For example: Time Jun UR (monthly) 6 UR (quarterly)

July Aug Sept Oct Nov Time Jun July Aug Sept Oct Nov

7 6 5 4 6 RGRT (monthly) 600 700 600 500 400 600

(6+7+6)/3 = 6.3

(5+4+6)/3 = 5 RGRT (quarterly)

(600+700+600) = 1900

(500+400+600) = 1500

-Follow the procedure in Excel Guide 6, starting with 10 lags, test to see whether RGRT, UR and AWE are stationary. When carrying out the tests, you also determine the optimal lag length for

each of the variables to be included in the final model. -Model specification: What kind of model is appropriate? Static, DL, AR, or ADL? If your variables are stationary, the final model may be

specified as RGRT = Lags of RGRT + UR + Lags of UR + AWE + Lags of AWE (Time Index and Seasonal Dummies may also be included). If the variables are non-stationary, you should use the following

model: ?RGRT = Lags of ?RGRT + ?UR + Lags of ?UR + ?AWE + Lags of ?AWE (again, Time Index and Seasonal Dummies may also be included). Note: When running regression, make sure that the Time

Index always starts with 1. For instance, if initially, you have data from Jan01 to Dec-01, Time Index for Jan-01 = 1. If later, you remove the observations for Jan and Feb, Time Index for Mar-01 = 1.

Empirical Results: -Run regression for your final model and produce the Table of Results. -Remove insignificant variables and re-run regression. -Don’t panic if many of the variables are

insignificant. Simply remove them all, except those found WITHIN the optimal lag length. For instance, for the results below, you should remove X1L4, X1L5, X2, X2L1, and X3.

Coefficients Intercept 54.1419723 X1L1 -0.1053366 X1L2 0.0119783 X1L3 0.60826034 X1L4 -0.0342699 X1L5 -0.1853314 X2 69.3895469 X2L1 17.2890504 X3 -0.0715066

Standard Error 20.43306823 0.143137397 0.139009138 0.130541361 0.138037475 0.136442807 43.02579734 43.42862986 0.577541435

t Stat 2.649723 -0.73591 0.086169 4.659522 -0.24827 -1.35831 1.612743 0.398103 -0.12381

Lower Upper P-value 95% 95% 0.010756 13.10095 95.183 0.465223 -0.39284 0.182163 0.931676 -0.26723 0.291186 2.38E-05 0.34606 0.87046 0.804946 -0.31153 0.242987 0.180463 -0.45938 0.088722

0.113095 -17.0303 155.8094 0.692249 -69.9399 104.518 0.90196 -1.23153 1.08852

-Utilising your regression model, forecast RGRT. It is sufficient to forecast for the first quarter of 2014; you don’t have to forecast for the whole year. An example: Suppose you have the

following data:
Time Jan-01 Feb-01 Mar-01 Apr-01 May-01 Jun-01 Jul-01 Aug-01 Sales 308.6 313.9 302.5 323 315.3 320.4 339.4 362.7 Income 200 210 190 210 209 212 218 220 5.3 -11.4 20.5 -7.7 5.1 19 23.3 5.3 -11.4 20.5

-7.7 5.1 19 5.3 -11.4 20.5 -7.7 5.1 5.3 -11.4 20.5 -7.7 10 -20 20 -1 3 6 2 10 -20 20 -1 3 6 DSales DSales L1 DSales L2 DSales L3 Dincom e DIncome L1

Sep-01 Oct-01 Nov-01 Dec-01 Jan-02 Feb-02 Mar-02 Apr-02 May-02

330.5 340.8 367.7 380.2 373.38 379.36 367.52 388.02

213 215 221 225 224 223 220 230

-32.2 10.3 26.9 12.5 -6.82 5.98 -11.84 20.5

23.3 -32.2 10.3 26.9 12.5 -6.82 5.98 -11.84 20.5

19 23.3 -32.2 10.3 26.9 12.5 -6.82 5.98 -11.84 20.5

5.1 19 23.3 -32.2 10.3 26.9 12.5 -6.82 5.98 -11.84 20.5

-7 2 6 4 -1 -1 -3 10

2 -7 2 6 4 -1 -1 -3 10

And, after removing insignificant variables, the equation of relationship is: ?Sales = 3.21 – 0.33?SalesL1 +1.65?SalesL2 + 0.78?SalesL3 For May-02, forecast ?Sales = 3.21 – 0.33*20.5 + 1.65*(-11.84) +

0.78*(5.98) = -18.42. However, you are asked to forecast Sales, not ?Sales Time Apr-02 May-02 Actual Sales 388.02 -18.42 388.02 – 18.42 = 369.6 Forecast ?Sales Forecast Sales

Note: You don’t have to interpret coefficients as in Assignment 1. Just forecast. Conclusions and Limitations: Summarise your findings and discuss the drawbacks of your report. References:

Harvard style.

Appendix: Include all necessary technical materials and data. The appendix should be well sorted out and with brief descriptions so that it is easy to follow.

PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT 🙂