Data & Methods
Data Collection
To address the overarching problem and the proposed research questions, our group opted to utilize the data set entitled “Consumer Price Index for All Income Households by Commodity Group (2018=100): January 2018 - March 2024” data set found on the OpenStat website , an initiative made by the Philippine Statistics Authority.
The data set as a whole is massive, as it lists the Consumer Price Index (CPI) of various commodities under different commodity groups, wherein the CPI is computed based on the province or region monthly.
Given the scope of our study, we will perform modifications onto the data set, which includes reformatting it as a whole followed by preprocessing. Details about our data set will be presented below.
About the data set
After performing preprocessing and reformatting, our data set now contains a total of 44850 entries. Moreover, the data set now has five columns instead of just two, which was found in the raw data set. From left to right, these columns are:
In this case, the years we consider are the years 2019 to 2023, since we specifically wanted to know the transportation situation in the past five years. It is worth noting that the year 2024 was not considered as data for this year is still incomplete. The months we consider are January to December with an additional average value. This aside, since every entry represents a totally different CPI value, no sampling method will be used here.
Exploratory Data Analysis (EDA)
Now that our data set is cleaned and well-formatted, we can now proceed to exploratory data analysis. Essentially, this step involves creating (clear) visualizations to help us understand the data, and using this information to determine if our established hypotheses are correct. Following this, we will perform hypothesis testing to see what results we can derive from the data.
Plot 1
What is the overall trend of the CPI of public transportation over the past five years?
From our visualization, we can see that the CPI of all transportation groups seem to be going upward over the years. Generally, the CPI for all transportation groups seem to be continuously increasing until it plateaus on some maximum and will eventually increase again after some amount of time.
Plot 2
Is there a relationship between the CPI of fuel/gas and public transportation over the past five years?
From here, we can see that the CPI for both fuel and land transportation seem to be going upwards, albeit at different rates. This suggests that they may be correlated to one another. Nevertheless, it is worth noting that the CPI for fuel fluctuates, whereas the CPI for land transportation is relatively more stable.
What is the overall trend of the CPI of public transportation over the past five years?
The Mann-Whitney U statistical test was then used to check if we will accept or reject the null hypothesis. After performing the statistical test, we obtained a p-value of 7.714x10^(-6), which is less than our set significance level of 0.05. Thus, we reject the null hypothesis.
Is there a relationship between the CPI of fuel/gas and public transportation over the past five years?
The Spearman Correlation test was then used to check if we will accept or reject the null hypothesis. After performing the statistical test, we obtained the p-value of 3.626x10^(-12), which is less than our set significance level of 0.05. Thus, we reject the null hypothesis. Additionally, the Spearman Coefficient rho of 0.7336 tells us that there is a positive correlation between the CPI of fuel and public transporation where both increase together over time.
