top of page

A Data Analytics Project: 
Assessing the possibility of recession in the U.S

Machine Learning and Statistical Data Analysis Analyzing and Categorizing Amazon Product R

The objective of this data analytics project is to assess the possibility of a recession in the United States by analyzing historical and future trends in economic indicators. The project involves several tasks: Cluster Analysis, Time Series Analysis and Time Series Forecasting

There has been a constant discussion on whether or not the U.S. is headed for a recession. There are various opinions and thoughts divided over it, many believe that these are mere fears. The National Bureau of Economic Research(NBER), who are the arbiter of recessions, haven’t declared any recession as yet. The technical definition of a recession is two consecutive quarters of declining gross domestic product(GDP), but U.S.’ NBER define a recession as widespread contraction in economic growth across various macroeconomic indicators, such as income, personal consumption, employment, and industrial production. Table below from Federal Reserve Bank shows the percentage change, on average, across recessions since 1947 for (a) the period prior to the recession and (b) the first two quarters subsequent to the peak.

Screenshot 2023-09-30 123152.png

Regional Analysis

With the conventional definition of a recession tied to GDP decrease, we analyzed the mean percentage of GDP across U.S. regions. The Mideast states showed the lowest increase at 0.27%, followed by 0.56% in the south and 0.67% in the Northeast. The west region exhibited the highest increase, likely due to major tech industry hubs. While this analysis hints at trends, clustering all states and macro-economic indicators using k-means would provide more insights to predict a 2023 recession.

Screenshot 2023-09-30 123409.png

K-Means Clustering Model

A dataset named "State.csv" was utilized for K-Means clustering, focusing on various variables and their quarterly percentages. Initially, a visualization depicted the geographical spread of variable values over time. After data cleaning and manipulation, K-Means clustering was performed, suggesting that 2 clusters would be optimal based on specific methods. Subsequently, the data was divided into these 2 clusters. The cluster plot showed that around 58% of the variability could be explained by the first two variable dimensions. Analyzing the mean values, negative percentage values for employment and wages in both clusters were observed, hinting at a possible recession. However, when considering other variables like GDP and income indicators, the data did not strongly indicate an imminent recession in the U.S.

Screenshot 2023-09-30 123918.png

Time Series Analysis

Time series analysis involves analyzing patterns and trends in data collected over time to make predictions about future values based on observed patterns. In the discussed graphs, a clear trend is observed in the time series over the years. Notably, there is a significant change at the end of 2020, with visible shifts in peaks and troughs. This abrupt change is attributed to the global pandemic that disrupted normal operations worldwide in 2020. Subsequently, efforts were made to stabilize and return to normal functioning, resulting in a more consistent pattern in the following two years (2021 and 2022) to prevent further decline.

Screenshot 2023-09-30 124632.png

Prediction Of Recession In FUTURE

Screenshot 2023-09-30 125429.png

Peer into the crystal ball of time series forecasting for the USA's economic indicators in 2023. Among them, income and manufacturing dance to a seasonal beat, while the others march in a stable rhythm akin to the last two quarters of the previous year. Glancing at the point forecast table, spotlight on Q4 2022, all indicators gleam with positivity, except manufacturing, which takes a slight dip at -0.0374%. Yet, fear not, for this lone negative ranger doesn't shout 'recession'! Some graphs show a gentle downward drift, perhaps swayed by recent pandemics and conflicts.

In sum, the forecast isn't sounding the recession alarms for the vibrant year that is 2023! 🌟📈

Screenshot 2023-09-30 125612.png

The economic indicators—income, employment, expenditure, production, and manufacturing—are like musical notes in predicting a recession. But why stop at a solo when we can orchestrate an ensemble of data for a symphony of insights? Let's zoom in, breaking down the data to a state-level, unveiling regional variations and vulnerabilities. Oh, and let's not forget the plot twists—new government policies, the housing market's heartbeat, and the pulse of consumer sentiment. These variables are the drumbeats and crescendos in our economic symphony. As of now, the grand finale rings clear: the USA is not in the recession spotlight for 2023! 🎵🎉

Complete Report


bottom of page