Count: 48

Task 5

Count: 48

SUM(Sold Price)

Sum: 111,856

Average: 2,330

Minimum: 115

Maximum: 24,500

Median: 990

Standard deviation: 4,156

First quartile: 581

Third quartile: 2,211

Skewness: 4.06

Excess Kurtosis: 17.26

a) The mean = 2,330 is greater than the median = 990, indicating that the sample data is positively skewed not normally distributed. The skewness value is also 4.06, further implying positive skewness. The excess kurtosis is equal to 17.26 which is large enough to conclude that the data was not obtained from a normally distributed population.

b) P(-1.5 < z < 1.5) = P(z<1.5) P(z<-1.5) = 0.9332 0.0668 = 0.8664

Hence, we expect 86.64 percent of the observations to be within 1.5 standard deviations of the mean. In count terms, that is 0.8664 * 48 = 42

c) x = zs + ux_lower = (-1.5 * 4156) + 2330 = -3904x_higher = (1.5 * 4156) + 2330 = 8564

Hence 46 out of the 48 observations in the sample or 95.83 percent actually fall within the 1.5 standard deviation from the mean.

d) The count obtained in b) is 42 while the actual count obtained in c) is 46. Hence, the count in b) is less than that of c). Hence, the conclusion matches well the visualization in a) that indeed the sample data is not normally distributed.

Task 6

Sold Price

Mean 2330.328

Standard Error 599.8422

Median 990.15

Mode #N/A

Standard Deviation 4155.829

Sample Variance 17270911

Kurtosis 19.338

Skewness 4.192262

Range 24385

Minimum 115

Maximum 24500

Sum 111855.8

Count 48

Confidence Level(90.0%) 1006.491

a)

i) point estimate of the mean sold price = 2330.328

ii) lower limit = 2330.328 1006.491 = 1323.837

upper limit = 2330.328 + 1006.491 = 3336.819

The 90% confidence interval is [1323.837, 3336.819].

iii) We are 90% confident that the true mean sold price is between 1323.837 and 3336.819.

iv) since the population standard deviation is unknown, we use the critical t

critical t (0.10/2, 47) = 1.68

standard error = 599.8422

margin of error = 1.68 * 599.8422 = 1007.7350

lower limit = 2330.328 - 1007.7350 = 1322.593

upper limit = 2330.328 + 1007.7350 = 3338.063

Hence, the 90% confidence interval is [1322.593, 3338.063].

b) Yes, the interval estimate is satisfactory because 1650 is contained within the 90% confidence interval.

Task 7 a)

Brick Veneer

Mean 0.38

Standard Error 0.069341

Median 0

Mode 0

Standard Deviation 0.490314

Sample Variance 0.240408

Kurtosis -1.81429

Skewness 0.509877

Range 1

Minimum 0

Maximum 1

Sum 19

Count 50

Confidence Level(99.0%) 0.18583

i) point estimate of the proportion of brick veneer properties = 0.38

ii) lower limit = 0.38 0.1858 = 0.1942

upper limit = 0.38 + 0.1858 = 0.5658

Hence, the 99% confidence interval estimate of the proportion of brick veneer properties is [0.1942, 0.5658].

b)

Use z-critical because this is estimating population proportion which always makes use of the value of z.

z-critical = 1.96

standard error = 0.07

margin of error = 1.96 * 0.07 = 0.1373

lower limit = 0.38 0.1373 = 0.2427

upper limit = 0.38 + 0.1373 = 0.5173

Hence, the 95% confidence interval estimate of the proportion of brick and veneer properties is [0.2427, 0.5173].

c) The interval is a)ii) is wider since it makes use of a higher confidence interval at 99% while the interval in b) is narrower since it makes use of a lower confidence interval at 95%. Hence, the higher the confidence interval, the higher the precision is and the wider the interval will be.

BEO1000 Business Data Analytics and Visualisation

2023

Assessment 4

Storytelling Presentation

(To be submitted online to VU Collaborate Dropbox before Workshop Session 10

Your name and student ID to be included in the file name)

_________________________________________________________________________

Scenario: At the end of your second year Bachelor of Business at VU Melbourne, you were offered an eight-week internship program at one of the Big Four accounting/consulting firms. After one week of induction, you were assigned to the financial service team to support a consulting project for a client who is seeking market intelligence. Your assigned task is specifically to research Melbournes current residential property market condition as part of the project. You were informed that you will be presenting your initial findings in the teams progress meeting with the client next week.

From the Business Data Analytics and Visualisation unit in the first year at VU you have learnt that a random sample would be the best representative of the population and projecting the population by a random sample is the most time-and-cost effective way to study the population. From the economics class in the first year, you also learnt that price is an indicator of the market. So you will focus your study on property prices for this project. You know that with strong theoretical support, the story will be more convincing, so you have browsed the internet and found an interesting property pricing model called hedonic pricing. Hedonic pricing models property price difference by external and internal factors of the property, holding constant of the market (supply and demand) condition. You proposed to use a representative sample of property prices to paint an overall picture of the property market in Melbourne and storytell the picture. With only a week of time to prepare for the presentation, to make the initial study manageable, you use a reduced form of the hedonic pricing model. If you successfully paint a vivid picture with the initial study using the reduced hedonic pricing model, there could be the potential to negotiate for more time and financial resource for a thorough study. Your team leader agreed with your proposal.

You collected a sample of 50 properties from around the Melbourne metropolitan area within a certain period of time, say within last week. The sample consists of information about the region, the property type, the size of the land, the number of rooms, the building type, the advertised price, the sold price and the sales results (that is the sample you have used for Assessment 3). You believed thatboth advertised price and sold price are affected by region, property type, building type, number of rooms and land size while the advertised price is the sellers expectation and the sold price is the realization of the market. Some of the properties are matching the sellers expectations with market realization well but others are not (that is market realisation exceeding or falling short of expectation). With your educated guess, you suspect that the sales result also have some influence on the difference between the advertised price and the sold price. You decided to be selective and are going to pick the most interesting story of what the data can tell.

When meeting a client and making a presentation, you are representing the firm you are working for. Not only is the data story you tell an indication of the quality the firms service, your appearance, voice, tone, facial expression, gesture and posture form part of the picture your client will judge of your firm. To improve your presentation, you decided to capture your practice presentation by video recording and ask your team members for their feedback.

Your Task

You will be using the Panopto video capturing technology made available from VU Collaborate to record a video including (1) your demonstration on the use of Tableau for data visualisation (and you may also demonstrate the use of Excel for data analytics if you are interested to do so), (2) the image of yourself throughout the demonstration and (3) your voice of verbal explanation of the data patterns and storytelling the findings. The following link is the instructions on How to Record a Video in a Web Browser (Panopto Capture).

You will be using the same sample of 50 observations of Melbourne property data you have used for your assignment in Assessment 3. Your data visualisation and storytelling will be focusing on the Advertised Price of the properties. You may be interested in explaining to your client about how building type influences the advertised price, or how the location of the property, or the land size or the number of rooms or any other factors have to do with the advertised price. Or you may be interest in how those factors jointly influence the advertise price difference (if so, in addition to using Tableau for data visual, you need to use Excel for data analytics). The choice is totally up to you.

At presentation, you should start with thanking your client to trust your firms service and thanking your team to give you the opportunity to present your data analysis. You should also briefly introduce yourself including your name, your tertiary course and data analysis is your chosen specialisation to establish your credentials.

Your allocated time for the presentation is between 8 to 10 minutes and you must have to stick with the rule of not going over the time limit for a business presentation. Practicing and recording your presentation a number of times and review your recordings and choose the one you are happy with for submission.

Submission

The video recording file (in MP4 format) will be submitted to the VU Collaborate Assessment 4 Dropbox before Workshop Session 10. The following is the link to the instruction on Submitting a Panopto Video into an Assessment Dropbox.

Only online submission to the dropbox can be marked and recorded. All files to be submitted to the dropbox should include your name and your student ID as part of the file names.

Assessment

The quality of your storytelling presentation will be judged by your confidence in public speaking, your efficiency in operating the Tableau software (and Excel if needed), your insight into the data patterns you have identified and the story exciting your client.

The full mark of the video presentation is 15 marks. A rubric for Assessment 4 is available on VU Collaborate for your reference.

Your teacher will demonstrate how to use Panopto video recording in Workshop session 8 and demonstrate a number of possible ways of using Tableau to generate interactive visuals and tell the story the data reveal in Workshop session 9.

P.S. For interested students, if you wish to investigate the joint effect of a number of variables such as Region, Building Type, Land Size and Number of Rooms on the Advertise Price, you can use the multiple regression methods. You are venturing into an advanced regression model building practice which belongs to an econometrics course. The process to generate multiple regression in Excel is similar to generating a simple regression we have discussed in Workshop sessions 8 and 9. Certain preparation on your sample data is needed in order use the Data/Data Analysis/Regression functions in Excel to accommodate categorical variables such as Region or Building Type. You should organize a time with your teacher to learn how to generate the set of dummy variables for the preparation for multiple regression. Note, be mindful that for each additional variable included in a regression equation will erode the explanatory power of the regression for certain degree. Since your sample size is 50 observations you must be cautious not to include too many dummy variables when building your multiple regression.

END OF PART ASSESSMENT 4.