Normal Distribution Analysis Of Utility Charges For One-Bedroom Apartments

PART A:
Suppose that the mean download time for a commercial tax preparation site is 2.0 seconds. Suppose that the download time is normally distributed, with a standard deviation of 0.5 second.  What is the probability that a download time is:

• above 1.8 seconds?
• between 1.5 and 2.5 seconds?
• 99% of the download times are slower (a higher number of seconds   taken to download) than how many seconds? 

PART B:
The file Utility contains the electricity costs, in dollars, during July 2010 for a random sample of 50 one-bedroom apartments in a large city. Decide whether the data appear to be approximately normally distributed. 

Decide whether the data appear to be approximately normally distributed by:

(a) Constructing a box plot.

(b) Constructing a histogram.

(c) Comparing data characteristics to theoretical properties.

(d) Constructing a Quantile-Quantile Normal Probability Plot.

You must include all Excel results for this data as part of the justification for your conclusions. (Include them within the appropriate section – not at the end of your document.)

Your assignment must be submitted as a single word document via LMS.      You will find the data in Utility in LMS under Topic 5, Data Files.¹ Based on Question 6.22 in Levine et al (2013) 

Further comments:

Your completed assignment needs to include the following components:

PART A: Completed questions.
PART B:

• An Introduction, where the key aspects of this topic are fully discussed. (This discussion should be referenced using the Chicago referencing method)
• The Body of your assignment, where all of your findings are presented. (This should include relevant Excel results, graphs, tables…) Each of the pieces of information that are included need to be discussed in relation to the given question.
• A Conclusion, which summarises your key findings and links these back to your comments in the introduction.
• A Bibliography which includes all resources referenced within your assignment.

INCLUDE THE MARKING GUIDE (BELOW) AT THE END OF YOUR ASSIGNMENT. 

Marks will be allocated according to the quality of each of the components above. It is important that you present your assignment with sufficient detail to fully answer the given question, while at the same time present your arguments in a concise way. (As a guide: Word Limit, at least 1500 words). 

IMPORTANT NOTE: Remember that this is an individual assignment. You must write the discussion sections of the assignment yourself and in your own words. Do NOT simply replace words from another student’s work with similar words. All assignments which include plagiarism of any form will be officially reported for further investigation.

NAME:

PART A                                                                                  MARK AND COMMENT

 PART B                                                                                    (a)                                                                                            MARK AND COMMENT (b)                                                                                          MARK AND COMMENT(c)                                                                                            MARK AND COMMENT(d)                                                                                            MARK AND COMMENT   

(a)	/2
(b)	/3
(c)	/3
Box Plot: -Plot, Title, Label	/2
- Five Number Summary	/1
Summary Comments on Box plot	/2
Histogram: -Title -Label	/2
-No gaps -Corrected bin values	/2
-Suitable bin size	/1
Summary Comments on Histogram	/2
Theoretical Properties: Included Summary Stats	/2
Compared Mean and Median	/1
IQR ~ 1.33 times the std dev	/1
Range ~ 6 times std dev	/1
68.26% values lie +/- 1 std dev of mean	/1
80% values lie +/- 1.28 std dev of mean	/1
95.44% values lie +/- 2 std dev of mean	/1
Skewness zero	/1
Kurtosis zero	/1
Summary Comments on Theoretical Properties	/2
Normal Probability Plot	/2
Table of Values: - Ordered X - Ordered Probability - Ordered Z	/3
Summary Comments on Normal Probability Plot	/2
INTRODUCTION CONCLUSION BIBLIOGRAPHY	/3 /3 /1
BONUS POINTS: TOTAL:	/46

The average download time is 2 seconds and a standard deviation of 0.5 seconds.

1. The probability that the download is above 1.8 seconds?

The probability associated with the z-score value of -0.4 is less than 0.001. Therefore, the probability of the download time being above 1.8 seconds is above 0.999.

1. The probability that the download is between 1.5 and 2.5 

The probability of the download time is between 1.5 and 2.5 seconds is

1. 99% of the download times are slower than how many seconds?

PART B  Introduction

Normal distribution is among the most used distributions in statistics. Their distribution can be in different formats based on the mean and standard deviation. These statistics determine the position and the shape. A widely distributed population has a normal distribution which has a large kurtosis and longer tails. In addition, a population whose values are not approximately distributed around the mean has a skewness statistic which is not equivalent to zero[1].

A negative and positive skew indicates that the left and right tail are longer respectively – hence the mean and median statistics are not equal. Therefore, using lower and higher moments of statistics, the distribution of data can be effectively described. In a normally distributed population, respective probability values can be estimated using the standard normal and inverse standard normal tables[2]. There are various theoretical assumptions which can be used in determining whether a sample is skewed or approximately normal. Also, boxplots, histograms and QQ-plots can be used to check normality. In this paper, we will use the standard normal table, boxplot, histogram and QQ-plot to check for normality of utility charges (electricity costs) of one-bedroom units in a certain city.

Results & Discussions

1. Constructing Boxplot

Table 1: Summary statistics of Utility charges (Electricity costs)

 [1] Abdal-sahib et al., “Testing for Normality.”[2] Fletcher, “Normal Distribution.”        Figure 1: Boxplot of Electricity costs in dollarsAccording to figure 1 above, utility charges on electricity use seems to be approximately normal because the median approximately divides the dataset into two equal sections. In addition, the data does not have extreme values which could be affecting the distribution of the electricity charges. Therefore, as it has been observed in the summary statistics – mean and the median values are close to each other, the boxplot affirms the data is approximately normally distributed.

• Constructing Histogram

Table2: Histogram frequency table  Figure 2: Unsorted Histogram of Utility Charges (Electricity costs)

Figure 2 above is a histogram chart with values of the electricity costs for the people in the city. The histogram was plotted by using bins with differences of 10 from 80 to 220. The selection of the bins was guided by the minimum and the maximum values. Most of the electricity bills were between 140 and 140 dollars which had a frequency of 8 one-bedroomed apartments. According to the distribution of the frequency in the histogram table and chart above, we can confirm that the data is approximately equally distributed around the arithmetic mean[1].  [1] Scott, “Histogram.”

 

 Figure 3: Sorted Histogram with Cumulative frequencies

The sorted histogram above represents the bins in a chart from the one with the highest frequency to the least. One-bedroom units in the 150, 130, 160, 170, 120 and 180 categories cover 56% of the entire sample. 220 category had the least frequency and 150 had the highest. 

1. Data Characteristics and Theoretical Properties

On average, the amount (in dollars) a resident in the city pays for the electricity is 147.06 dollars with a standard deviation of 31.691 dollars. There is no much difference between the average value and the median, indicating that the data is not highly skewed, hence approximately normally distributed. Based on the skewness statistics which is 0.016, we can conclude the data is slightly skewed to the right. The kurtosis is less than three hence concluding that the distribution has less extreme values as compared to the standard normal distribution, hence the peak is sharper. An individual who pays the highest amount of electricity cost in the city pays around 213 dollars. In the same way, the least amount paid based on the sample was 82 dollars.

The interquartile range is 42.75, which is approximately equal to the theoretical assumption that it should be 1.33 times the standard deviation (42.15). The range statistic for the one-bedroom electricity costs is 131 and the theoretical assumption that it should be 6 times the standard deviation has not been met (6 times of standard deviation is 190.146). There are 34 values between negative and positive 1 standard score, which translate to 68%, hence the theoretical assumption is met. 40 values lie between negative and positive 1.28 which is 80% as defined in the normality theory. There are 48 values between negative and positive 2 which translates to 96%, hence the confirmation according to standard normal assumptions. 

1. Quantile-Quantile Normal Probability Plot

Table 3: Cumulative Distribution Function and Standardized Values

Utility Charge (Electricity Costs)
Mean	147.06
Standard Error	4.482
Median	148.5
Mode	130
Standard Deviation	31.691
Sample Variance	1004.343
Kurtosis	-0.544
Skewness	0.016
Range	131
Minimum	82
1st quartile	126
3rd Quartile	168.75
Maximum	213
Sum	7353
Count	50
Unsorted Bin	Frequency	Cumulative %	Sorted Bin	Frequency	Cumulative %
80	0	0%	150	8	16%
90	2	4%	130	6	28%
100	2	8%	160	5	38%
110	3	14%	170	5	48%
120	4	22%	120	4	56%
130	6	34%	180	4	64%
140	3	40%	110	3	70%
150	8	56%	140	3	76%
160	5	66%	190	3	82%
170	5	76%	90	2	86%
180	4	84%	100	2	90%
190	3	90%	200	2	94%
200	2	94%	210	2	98%
210	2	98%	220	1	100%
220	1	100%	80	0	100%
More	0	100%	More	0	100%

i	Utility Charge	CDF	Z- Score	Standardized Values
1	82	0.01	-2.32635	-65.06
2	90	0.03	-1.88079	-57.06
3	95	0.05	-1.64485	-52.06
4	96	0.07	-1.47579	-51.06
5	102	0.09	-1.34076	-45.06
6	108	0.11	-1.22653	-39.06
7	109	0.13	-1.12639	-38.06
8	111	0.15	-1.03643	-36.06
9	114	0.17	-0.95417	-33.06
10	116	0.19	-0.87790	-31.06
11	119	0.21	-0.80642	-28.06
12	123	0.23	-0.73885	-24.06
13	127	0.25	-0.67449	-20.06
14	128	0.27	-0.61281	-19.06
15	129	0.29	-0.55338	-18.06
16	130	0.31	-0.49585	-17.06
17	130	0.33	-0.43991	-17.06
18	135	0.35	-0.38532	-12.06
19	137	0.37	-0.33185	-10.06
20	139	0.39	-0.27932	-8.06
21	141	0.41	-0.22754	-6.06
22	143	0.43	-0.17637	-4.06
23	144	0.45	-0.12566	-3.06
24	147	0.47	-0.07527	-0.06
25	148	0.49	-0.02507	0.94
26	149	0.51	0.02507	1.94
27	149	0.53	0.07527	1.94
28	150	0.55	0.12566	2.94
29	151	0.57	0.17637	3.94
30	153	0.59	0.22754	5.94
31	154	0.61	0.27932	6.94
32	157	0.63	0.33185	9.94
33	158	0.65	0.38532	10.94
34	163	0.67	0.43991	15.94
35	165	0.69	0.49585	17.94
36	166	0.71	0.55338	18.94
37	167	0.73	0.61281	19.94
38	168	0.75	0.67449	20.94
39	171	0.77	0.73885	23.94
40	172	0.79	0.80642	24.94
41	175	0.81	0.87790	27.94
42	178	0.83	0.95417	30.94
43	183	0.85	1.03643	35.94
44	185	0.87	1.12639	37.94
45	187	0.89	1.22653	39.94
46	191	0.91	1.34076	43.94
47	197	0.93	1.47579	49.94
48	202	0.95	1.64485	54.94
49	206	0.97	1.88079	58.94
50	213	0.99	2.32635	65.94

Fletcher, “Normal Distribution.” 

Figure 4: QQ Plot of Utility Charges (Electricity Costs)

In the development of the QQ-plot in figure 4 above, cumulative probabilities, z-scores and standardized values were calculated and represented in the plot. The x-axis represents the z-scores and the y-axis is the standardized values for the electricity costs. According to the QQ-plot above, the standardized values seems to be following a straight line, hence the conclusion that the data is approximately normally distributed[1]. It can also be observed that most of the standard values are concentrated around the standard mean, which is zero. In addition, the deviations (standard values) from the mean are approximately distributed equally from the mean, hence proving that the data is not skewed or rather if it is skewed, the skewness does not affect the distribution.

Conclusion

In conclusion, usage of electricity on the city in one-bedroom apartments is approximately normally distributed with no extreme values. Most of the house units (one-bedroom) pay for electricity between 120 to 200 dollars. Finally, usage of electricity in the city for one-bedroomed apartments does not differ extensively from one unit to another and the distribution produces less extreme values (outliers) compared to the ideal standard normal. Using standard normal tables, histogram, and boxplot, we conclude that the data is normally distributed. Therefore, the results from these methods are homogenous. 

Abdal-sahib, Reem, Shahah Musaed Altammar, Haffiezhah An-nadiah Azlan, Ali Aytemur, Stephanie Balters, Martin Steinert, Christo A. Bisschoff, et al. “Testing for Normality.” Frontiers in Psychology 98, no. 1 (2013): 1–8. 

Fletcher, J. “Normal Distribution.” BMJ 338, no. feb18 2 (February 18, 2009): b646–b646. 

Oldford, R. Wayne. “Self-Calibrating Quantile–Quantile Plots.” American Statistician 70, no. 1 (2016): 74–90. 

Scott, David W. “Histogram.” Wiley Interdisciplinary Reviews: Computational Statistics 2, no. 1 (2010): 44–48. 

Oldford, “Self-Calibrating Quantile–Quantile Plots.”