代写Qbus6860 Group Assignment代做留学生SQL语言程序
- 首页 >> DatabaseQbus6860 Group Assignment
Introduction
The Consumer Price Index (CPI) and Retail Turnover (RT) are two critical economic indicators. As a consulting firm, we leverage these two key economic indicators to offer valuable insights to enterprises. These indicators serve as essential tools for understanding customer consumption behavior, assessing industry conditions, and exploring broader economic trends. That information can assist companies' decision-making and help them to adapt to the ever-changing economic landscape.
In this report, we focus on CPI and RT data from different industries and states within Australia. Initially, we will explore the relationship between inflation rate and turnover. Additionally, the regional and industry-specific analysis offers a perspective on how economic conditions and consumer behavior vary across the country. However, the COVID-19 pandemic has introduced unique challenges, profoundly impacting economic conditions. The relationship between inflation rate and retailing turnover in different industries and states under the COVID-19 period is a complex interplay of factors. The pandemic has led to significant shifts in consumer demand, supply chain disruptions, and changes in consumption patterns. Indeed, these changes have exhibited distinct patterns across various industries and states. In this report, we have identified and analyzed these variations. Of particular interest is the distinctive performance of the food industry in comparison to other sectors during both pandemic and non-pandemic periods. That makes us conduct further in-depth analysis.
Data processing
Our data downloads two datasets from data.gov.au, the Retail Turnover and Consumer Price Index (CPI) datasets. These datasets cover relevant information from 2010 to 2023. We noted that 317 turnover values were missing in the retail turnover data, while there were no missing core index values in the CPI data. We did not take immediate action on the missing data, and in our analysis, we were concerned with exploring the relationship between retail sales and the CPI. We decided first to merge the two data to facilitate the analysis of key business issues.
Firstly, we examined the structure of the two datasets and found that the two datasets used different industry classification criteria. To ensure comparability and consistency, we created an industry code transformation mapping so that the industry codes in the CPI dataset corresponded to those in the retail sales dataset. It is worth noting that there is no appropriate classification for Cafes, restaurants and takeaway food services in the CPI data, and we decided to merge this classification into the food retailing. Secondly, we noticed missing or inconsistent values in the CPI data during the merge. Therefore, we created the delete_repeat_data function, which not only removes records with empty CPI values but also calculates the average of the data over the same period, industry, and region by group aggregation. Finally, we filtered the variables and renamed the two datasets to ensure the column names were consistent when merging. Once the data were prepared, we integrated the retail sales and CPI datasets into a new data frame, merge_data, by connecting the keys standard to both data - time period, industry, and region. We deal with the missing value by filling the mean, because directly dropping null and filling forward will cause bias, compared with the original data set. Ultimately, we computed the percentage change in retail sales versus the CPI on the merged dataset. This step lays the foundation for subsequent insights into retail turnover and CPI changes over time across industries and regions.
Analysis
1.0 Overview of relationship between retail turnover and inflation
To analyze the relationship between retail turnover and inflation in Australia, we first created a line plot to examine the trends of these two variables over time. We excluded Australia from regions, then grouped the data by time period and calculated mean values of retail turnover and inflation rate for each time period. Due to the extreme value of inflation rate at the 2020-Q3 time point, we chose to drop the data for this point to better visualize the graph. This operation does not affect our overall trend analysis.
Retail turnover displays an overall pattern of fluctuating growth. Over the course of 13 years, it has risen from an initial value of about 1000 million AUD to 1700 million AUD. For the inflation rate, it exhibits relatively significant fluctuations, with most of the data falling within the range of 0 to 0.6. The upward trend in inflation rate values is continuous in most time periods, although there may be slight fluctuations within certain quarters. This suggests that the overall economy experiences a relatively sustained increase in prices. Notably, starting from the beginning of 2020, both retail turnover and inflation rate have experienced significant fluctuations. They both experienced a decline, especially with CPI witnessing a significant decrease. These fluctuations may be attributed to the impact of the COVID-19 pandemic and related economic factors. However, in this graph, apart from the overall trend, we cannot clearly discern the correlation between the inflation rate and retail turnover, and further analysis is needed.
1.1 The lagged cross-correlation between retail turnover and inflation
Next, analyze the lagged cross-correlation between retail turnover and inflation. Since the dataset covers a period from the first quarter of 2010 to the second quarter of 2023, totaling 13 years or 52 quarters, we chose to investigate the association between these two variables over the past 15 quarters and the next 15 quarters, which is a relatively long span. We used a loop to iterate through each lag value. Then we calculated the correlation between the percentage change of turnover and the lagged inflation.
The direction and strength of the correlation values for both positive and negative lags provide valuable insights into the connection between retail turnover and inflation. When the lag is set to 0, it means that retail turnover and inflation change simultaneously within the same quarter. A positive correlation suggests a positive relationship between them and vice versa. The magnitude of the correlation coefficient indicates the strength of the relationship.
Larger positive or negative values indicate a stronger relationship. Negative lags suggest that the inflation rate leads to retail turnover, while positive lags suggest the opposite.
According to appendix 1, it shows that at Lag 0, a correlation of -0.0711 indicates a negative relationship between inflation rate and retail turnover, implying that they exhibit opposite trends during the same period. However, the magnitude of this correlation is not substantial. It's important to note that this negative correlation is observed only at lag 0, and it may be influenced by seasonality or other factors. Moreover, the data utilized in this analysis is recorded on a quarterly basis, potentially obscuring short-term fluctuations, and dynamics. Thus, the actual relationship between inflation rate and retail turnover may vary over time.
At lag -6, the correlation value is 0.0478. It is important to note that this correlation coefficient reaches its peak within the overall data. This observation indicates that during the period corresponding to lag -6, there is a positive correlation between inflation rate and retail turnover. Furthermore, it suggests that inflation rate leads retail turnover within the earlier six quarters, meaning that changes in inflation six quarters ahead of retail turnover have a positive influence on future retail turnover. At other lag periods, the magnitude of correlation coefficient values are relatively small, indicating a relatively weak negative correlation between inflation and retail turnover. In conclusion, based on the chart, we can infer that changes in inflation occurring six quarters prior to retail turnover have a positive impact on future retail turnover.
2.0 The relationship between Mean Turnover and Different Regions
To compare the mean revenue generated over time in various regions, this report utilizes a bar chart. The x-axis represents the different regions, while the y-axis represents the mean turnover values (million AUD). On average over time, the New South Wales (NSW) are the highest retailing turnover. The possible reason for this is NSW has the highest population of any state in Australia, making it a significant consumer market (Channels, 2020). Moreover, NSW is Australia's largest state economy, and its economic diversity and emphasis on service-driven industries contribute to the high retail turnover (Channels, 2020).
2.1 The relationship between Inflation rate and Industry over time
The CPI provides insights into changes over time, and retail turnover refers to the total sales or revenue generated by a retail business over a specific period. Thus, this report can utilize a line chart to depict the time series changes in different industries, allowing for a comprehensive view of how various sectors change over time.
In the inflation rate chart, where the x-axis represents quarterly time periods and the y-axis represents inflation rate, with different colors representing various industries. Notably, compared with other industries, the department store industry between 2013 Q1 to 2021 Q1 has consistently maintained a slightly higher inflation rate. Despite fluctuations in the data, the overall average inflation rate for department stores remains slightly higher than that of other industries in that time period. This situation can be attributed to the classification of alcohol and tobacco belonging to the department store category.
The tobacco tax in Australia is among the highest in the world, reaching 65% in 2020 (Nicholas & Kelly, 2023). This high tax rate plays a significant role in driving up the cost of tobacco products and likely contributed to the increase in the CPI within the department store industry and subsequently affecting the inflation rate. Furthermore, to mitigate alcohol-related harms, the Australian government imposes taxes on alcohol (Alcohol and Drug Foundation: Position Paper Alcohol Taxation, 2023). These taxes are adjusted for inflation every six months in line with the CPI. That is probably one of the reasons for six months of lagged cross-correlation between retail turnover and inflation, which is mentioned in part one. This demonstrates the changes in taxation policies, especially in industries like alcohol and tobacco, can directly impact on inflation rate trends, and then influence the broader economy. After 2021 Q1, the inflation rate data begin to show different fluctuations, which can be attributed to the impact of the COVID-19 pandemic.
2.2 The relationship between Retailing Turnover and Industry over time
In the RT chart, where the x-axis represents quarterly time periods and the y-axis represents RT values (million AUD), with different colors indicating various industries. The food industry is the leading contributor to retail turnover. To explore more about the relationship between retailing turnover and industry, we used the retailing turnover data that had not been merged. The percentages of different industries in different states are calculated, as shown in
Appendix 2. It shows that the percentage of food retailing in each region is the highest, which is the same as the result obtained by using merge data.
Prior to the first quarter of 2019, retail turnover displayed a stable increase with a discernible pattern. However, after 2019 Q1, fluctuation appears in each industry. The reason for this change in behavior is the impact of the COVID-19 pandemic, which influenced consumer behavior and retail sales (Mizen, 2021). Since the onset of the pandemic, retail sales have experienced declines due to prolonged lockdowns and stay-at-home orders. These restrictions constrained spending and disrupted traditional shopping habits, leading to fluctuations in retail turnover data for various industries. The COVID-19 pandemic has had a profound and far-reaching effect on the retail sector, causing shifts in consumer preferences and purchasing patterns.
2.3 Correlation between Retail Turnover and Inflation Rate
The heatmap provides an overview of the correlation between Retail Turnover and Inflation Rate for different regions and industries. The x-axis represents various industries, and the y-axis represents different regions. Different colors indicate different levels of correlation.
One key observation is that food retailing exhibits the correlation between turnover and inflation rate is close to 0. That means regardless of whether the price of food retailing increases or not, people tend to continue buying. Additionally, this correlation pattern appears consistent across different regions, indicating customer behaviors and purchase patterns in food retailing will not be affected by different regions' culture, population and economy. On the other hand, the Department store has a negative correlation between turnover and inflation rate, possibly because alcohol and tobacco are expensive in Australia and not the necessary stuff for living, so people tend to reduce spending on them when prices rise. Another interesting finding is that Tasmania shows a significant negative correlation in other retailing compared to other states. In data processing, the "Other Retailing" category of RT includes Health, Education, Insurance, Financial Services, and others from the CPI. This suggests that residents in Tasmania have different spending behavior. in these industries compared with other states.
3.0 Overview of Covid-19
The epidemic has put unprecedented pressure on the global economy, with countries and governments facing significant economic challenges (Supple & Yu, 2023). Particularly in the retail sector (Olanrewaju & McSharry, 2022), the pandemic had a severe impact on retail sales due to travel restrictions. In addition, noted that COVID-19 also had a significant impact on the Consumer Price Index (CPI).
3.1 Turnover and inflation during the Covid period
Firstly, we control for the time variable referring to the time-series line graphs in the first part. Focusing on 2019 onwards, it is clear that from the beginning of 2020, the first quarter of 2020, there is a significant decline in turnover and inflation, with a more pronounced decline in inflation. The folded trend in inflation reflects the sharp volatility in 2020. Price movements in 2020 were heavily influenced by the COVID-19 pandemic, leading to increased volatility in headline inflation data (Reserve Bank of Australia, 2021). From the title of the chart and the red line markers, we can surmise that the COVID-19 pandemic may have been a major contributor to the plunge in inflation and retail turnover in early 2020. The outbreak may have contributed to shop closures, a drop in consumer confidence, and a slowdown in overall economic activity (Marsh, 2020). As can be seen from the graph, both are on an upward trend in the latter half of 2021, which may result from the beginning of economic recovery, restored consumer confidence and the economic stimulus measures taken by the government.
We can then see a pair of scrolling graphs of the correlation between retail turnover and inflation over time, which we divide into two periods: before and after COVID-19. The orange broken line indicates the strong and positive correlation between retail turnover and inflation before COVID-19. This means that when retail turnover increases, inflation also increases. At the beginning of 2020, the correlation dropped suddenly and sharply to almost zero, possibly due to fluctuations caused by the early stages ofthe COVID-19 pandemic. The blue dashed line indicates a sharp increase in the correlation from late 2020 to early 2021, followed by another sharp decline in the second half of 2021. This may be related to fluctuations in the COVID-19 epidemic and changes in government policy. However, when we explore this issue in depth, it is clear that retail turnover and inflation interact and possess a certain lag. When inflation rises, consumers will spend less if retail outlets sell non-essential goods. Inflation may reduce demand as consumers reallocate their spending to more essential goods and services. Conversely, a fall in inflation will likely stimulate consumers to spend (Amoussou, 2022).
3.2 Turnover and inflation during the Covid period in different regions and industry
In this section, we break down the impact of the COVID-19 epidemic on the relationship between retail turnover and inflation specific to different industries and regions of Australia. In this interactive bar chart, the x-axis is the region, the y-axis is the correlation between retail turnover and inflation, the blue bar is before the epidemic, and the red bar is at the time of the 2020 epidemic. For the total industry, the correlation is positive except in SA, where the two variables are negatively correlated before the epidemic.
We begin by looking at the FOOD RETAILING sector. The relationship between retail turnover and inflation in each region reversed from a positive to a negative correlation both before and during the epidemic, with the highest being more than -0.06 in WA. The decline in inflation during the epidemic, coupled with the fact that more and more Australians are cooking at home, supermarket and grocery shop sales increased dramatically (Commission Factory, 2022). Additionally, inflation began to rise in mid-2020 as a result of the consolidation of the food retail, restaurant and takeaway classifications, and the sharp downward trend in turnover was further exacerbated by the closure of a large number of restaurants as a result of the nationwide embargo, which severely affected shopping behavior (Commission Factory, 2022). Another hypothesis is that Inflation declines, leading to a significant decrease in food prices, a sharp increase in turnover, and an epidemic of residents stocking up on food.
And for household goods, a significant increase in positive correlation can be found. COVID-19 The pandemic has created an unprecedented demand for goods such as electronics, furniture, and sports equipment while people are cooped up in their homes (Amoussou, 2022). This can be found most dramatically in the correlation differentials for South Australia, Queensland and New South Wales. Tasmania, on the other hand, has seen little correlation or change in its correlation and may not have seen much change in the demand for household goods based on its geography.
The impact of the epidemic on the clothing industry is theoretically huge; inflation usually means higher prices, and when prices rise, consumers have less purchasing power. Suppose consumers expect inflation to continue to rise in the future. In that case, they may cut back on non-essential spending, which can lead to a drop in turnover in certain retail sectors, such as clothing and accessories (Brydges et al., 2021). However, we see exceptions in Victoria and WA, particularly in the WA region, where the correlation between turnover and inflation is significantly higher during the epidemic. This may be due to the various economic stimulus packages introduced by the WA government during this time. The state government's stimulus reportedly totalled $5.5 billion (Shepherd & Piesse, 2020).
The withering of the Department stores industry is visible to the naked eye. Especially in Victoria, the correlation changed from less than -0.05 to close to -0.25, with the negative correlation getting stronger. Multiple embargoes have led to a decline in consumer confidence and a reluctance to spend money on non-essential items (Mizen, 2021). As inflation has risen, as a reference point, Victoria is also "number one" in terms of retail sales declining by more than 10 per cent (Australian Bureau of Statistics, 2020). However, the negative correlation before and after the epidemics in South Australia and Western Australia has narrowed slightly.
Finally, let's turn our attention to Other retailing. There is a significant improvement in turnover and inflation correlation across all regions, especially in Queensland, from over -0.2 to a peak positive correlation of 0.32, but because our data for this industry includes a multitude of industry classifications, it also makes it impossible for us to be any more detailed in our specific analyses, which is also one of the shortcomings in our data analysis.
4.0 The insight for food industry
Upon conducting an analysis of various factors that influence the correlation between inflation and retail turnover, it emerged that the food industry persistently dominates in retail sales. During the COVID-19 pandemic, when most industries experienced substantial growth or decline, the food industry showed the same volatile trends as before the pandemic. This phenomenon aroused our interest: Why was the Australian food industry not affected by the epidemic crisis?
The horizontal axis of the graph represents time, spanning from 2010 to 2023, with the red dotted line marking the commencement of the COVID-19 pandemic. The vertical axis measures retail turnover in millions of Australian dollars. The graph's legend differentiates the sales trends across various industries through five distinct lines.
From the graph, it's apparent that before the pandemic, Australia's retail sales showed a clear seasonal trend during the second and fourth quarters of each year. This trend coincided with major holiday cycles, particularly Easter in the second quarter and Christmas in the fourth quarter. Retailers traditionally engage in promotional sales and marketing during these periods, offering significant discounts to stimulate consumer spending.
It can be seen that even after the outbreak, while many other industries experienced fluctuations in retail sales, the food industry's sales (the orange line) continued to grow steadily. This suggests that the outbreak did not have a negative impact on the food industry. This is due to the nature of food as a basic necessity, for which demand continues even during economic turmoil. In addition to this, the trend of home cooking during the epidemic, the panic-hoarding behavior of consumers, the rise of e-commerce, and supportive government policies have boosted consumer demand for food. In addition, steady export demand for 70 percent of Australia's agricultural products in international markets, even when hit hard by the epidemic, may also be a key factor underpinning solid food growth, according to the survey.
4.1 The relationship between CPI and RT for food sales
To explore the reasons behind this phenomenon, we analyzed the relationship between CPI and RT for food sales, and detail provided in appendix 3. It indicates that despite seasonal variations and market fluctuations, food industry prices and revenue have remained relatively stable, maintaining a consistent range for the most part. This is consistent with the conclusion obtained in part two that the relationship between inflation rate and RT is steadily approaching zero in different regions. The reason is that food is a fundamental necessity, with substantial consumer demand for it. Even if prices increase, individuals will continue to purchase them. It implies the food industry's remarkable resilience to economic fluctuations. Food expenditures by consumers have remained consistent despite the economic downturn and elevated inflation. Moreover, the spread of data points suggests that market competition, consumer preferences, and promotional strategies may affect retail food sales.
Shortcoming
1.0 The shortcoming of data
At first, in the process of merging datasets, due to the lack of consistency in industry standard classifications between the two datasets, we attempted to map industry standards from one dataset to another. However, there were some the industry names could not match. In such cases, we used subjective judgment to determine similar industry classifications. This approach could lead to inconsistencies in the analysis, which could impact the analysis of variables later, particularly when comparing different times or regions. We attempted to fill in the large number of missing values in the merged dataset with mean values. However, this approach may introduce bias as the data contains seasonality or a discernible trend. Furthermore, outliers could signify significant economic occurrences or natural phenomena. Eliminating or substituting the outliers could have an adverse impact on our analyses. Therefore, our choice was to maintain them.
2.0 The shortcoming of analysis
When analyzing our key business questions, the findings were based on correlation. However, correlation only indicates a certain degree of association between variables and cannot prove that a change in one variable is the cause of a change in another. Therefore, for the first question, we cannot conclude a causal relationship between the two variables. Due to different industry standards, we had to use our best judgment to find similar industry classifications. This could cause our analysis results to be inconsistent. Furthermore, the category 'Other Retailing' encompasses a wide range of industries, leaving us unable to discern which specific industries have influenced the relationship between the two variables. Also, when looking at how the pandemic changed the relationship between CPI and RT, it was said that government economic stimulus during the pandemic could affect retail sales and the consumer price index. However, we do not have sufficient information to see how these policies changed the connection between retail sales and inflation. Additionally, external factors such as global economic fluctuations and supply chain disruptions were not fully considered.
Conclusion
In conclusion, this report analyzed the relationship between the Inflation Rate and RT in Australia, emphasizing the disruptions caused by the COVID-19 pandemic. Our analysis revealed significant variations in this relationship across different industries and periods. Despite some data limitations and the challenge of inconsistent industry classifications, we found a complex interplay between CPI and RT, particularly in the food industry during the pandemic. For future research, more data may need to be considered such as actual data affected by economic activities to provide clearer insights into the economic impacts of such global events.
