I carried out a comprehensive analysis with the goal of highlighting important facets of organizational dynamics by examining pay scales by department, analyzing overtime patterns, and examine budgetary distributions. We will look at the effect of educational background on income. I also intend to find anomalies and outliers, offering insights into departmental and individual performance.
One statistical technique to determine if a sample’s mean deviates significantly from an assumed or known population mean is the Z-test. The Z-test, which is used in hypothesis testing, entails figuring out the Z-score, which expresses how far a data point deviates from the mean in standard deviations. An elevated Z-score indicates a noteworthy departure from the average. When the population standard deviation is known, this test is very helpful. The Z-test is a widely used statistical tool in many disciplines, including psychology, economics, and quality control.
We basically discussed the type of Dataset we should work on for our project along with that also discussed various Time Series Algorithms.
Time series are broken down into their component parts using conventional techniques like Autoregressive Integrated Moving Average (ARIMA) and Seasonal-Trend decomposition using LOESS (STL). ETS models, or exponential smoothing state space models, account for seasonality, trend, and error. Complex dependencies are handled by sophisticated machine learning models such as XGBoost, Gated Recurrent Units (GRU), and Long Short-Term Memory (LSTM). Seasonality-based forecasting is the area of expertise for algorithms like Prophet and SARIMA. Time series data trends and seasonality are accommodated by Holt-Winters exponential smoothing.
Studied Time series analysis, which is commonly used in fields such as finance and meteorology, involves examining data points ordered over time. By examining patterns, trends, and fluctuations in temporal data, this statistical method frequently provides insights into underlying behaviors. Finding trends, seasonal patterns, and autocorrelation are essential elements, and methods like decomposition and smoothing help reveal important information. In order to make predictions and identify anomalies, forecasting and anomaly detection are crucial components. Time series analysis helps to comprehend and utilize the temporal dependencies within data to make predictions and decisions by using techniques like Autoregressive Integrated Moving Average (ARIMA).
Analyzed the Dataset and this dataset includes economic indicators that the Boston Planning and Development Authority (BPDA) tracked on a monthly basis from January 2013 to December 2019. The data reflects the BPDA’s efforts to track and analyze important metrics for well-informed decision-making in city planning and development. These metrics span a variety of economic aspects, including employment, housing, travel, and real estate development for which analysis must be done further to generate insights from it.
We have finalized our report and are about to complete the material for the project with a recap of our findings.
Compiling the report along with making the advised changes by the team members specifically in the section of Discussion and Appendix A. Comparing the results of other members and trying to correctly explain the results and our findings in the convincing way possible.
Working on the draft of the report specifically on the sections of Findings and Discussion where my main finding with respect to the dataset must be explained in the concise way along with the Discussion that includes the fact that is proven by my findings.
Made some editing with respect to the project report and worked on trying to use other algorithms to get better accuracy for our data. Furthermore, discussed the insights among the group members to keep them all under one header. Along with that made some corrections upon input of the group members.
I started working on the report getting all the data results and writing report according to the described format. All the results are being added and a punchline report is being created to explain my results and findings.
