Fuel Prices PDF

Preview

Summary

Statistics Assignment based on Fuel Prices using the program R...

Description

Fuel Prices 1014SCG – Statistics Bree Preston-Miller s5177835 Individual Workshop Assignment 1 – 2019 Hamid 4pm – 6pm Mondays

Statistics 1014SCG

Introduction Fuel prices vary all over Brisbane with different variables having an influence on the price of the fuel, these include the different types of fuels, the brand being supermarket or non-supermarket and the location of the service station. A survey was conducted to obtain fuel prices of both diesel and unleaded at different times of the day from a supermarket brand and non-supermarket brand in six different areas of Brisbane. The data was collected with two different fuel checking websites due to the lack of transport to the service stations as well as the two websites do rely on the update from its customers to supply accurate fuel prices. To ensure that the fuel prices were accurate as possible, two websites were used, one being FuelMap Australia and the other being Petrol Spy.

Test of Proportion Z ¿

x−np √ npq

N = sample size = 162 P = predicted probability = 0.6 Q=p–1 = 1 – 0.6 = 0.4 Z = test of proportion X = How many expensive prices = 73

Z

¿

73−( 162 x 0.6 ) √ 162 x 0.6 x 0.4

Z = -3.881

Research Question Is the proportion of the diesel fuel prices that are considered to be high ( ≤ $145.9) more than 0.6?

Variables Variables are the price of the fuel and more specifically diesel.

1|Page Hamid 4pm-6pm Mondays

Bree Preston-Miller

s5177835

Statistics 1014SCG

Statistical Hypothesis H 0 : The proportion of the petrol prices that are more than $145.9 are more than 60% H 0 : pr(petrol prices being high) < 0.6 H 1 : The proportion of the petrol prices that are more than 64747 are not more than 60% H 1 : pr(petrol prices being high) > 0.6

Summary Statistics Graph 1: Histogram of Diesel Prices

Table 1: Statistics

Prices ($)

Minimum 139.9

Maximum 152.5

Mean 146.48

Standard Deviation 2.84

The price of diesel was predicted to have 60% of its values to be $145.9 or higher which is not supported by evidence so far. In table 1, it can be seen that the mean was 146.48 with the standard deviation of 2.84 allowing for the fact that most of the data set are between occurred between 144 and 148. This disagrees with the prediction as it is less than sixty percent is after 145.9. This can also be seen in graph 1, as a large portion of the data is placed within 144-145 while there is still a distribution of data more commonly between 146 to 152 compared to the distribution of data between 140-143.

2|Page Hamid 4pm-6pm Mondays

Bree Preston-Miller

s5177835

Statistics 1014SCG

Hypothesis Testing One Sample t-test

t = 0.01493, df = 161, p-value = 0.9881 alternative hypothesis: true mean is not equal to 146.48 95 percent confidence interval: 146.0424 146.9242 sample estimates: mean of x 146.4833 P-value is not between 1 – 5% which means the data set does not have strong evidence to support alternative hypothesis.

Fail to reject the Null hypothesis due to the lack strong evidence although a possibility of Type II error.

3|Page Hamid 4pm-6pm Mondays

Bree Preston-Miller

s5177835...