Order from us for quality, customized work in due time of your choice.
Project Topic: Time Series Analysis
1. You are given 2-year electricity consumption time series of an academic building in one of CUNY’s
colleges. The data file will be in CSV format, with hourly data granularity. You can also aggregate to
daily data if prefer.
2. The first task will be to cleanse the dataset before ana analysis. Typical data cleansing tasks include
removing or filling in missing values. Identifying and handling outliers and ensure data consistency.
3. Descriptive Analytics:
a. Calculating summary statistics such as mean, median, standard deviation, minimum, maximum,
etc., for the electricity consumption data. You may also compare the summary statistics for
electricity consumption before and during the COVID-19 Pandemic. This includes mean
consumption, standard deviation, minimum, maximum, and any other relevant metrics.
b. Create visualizations in Excel such as line charts or bar graphs to show the trend of electricity
usage over these different periods, highlighting any notable changes or patterns.
4. Predictive Analytics:
a. Time series trend analysis using regression or exponential smoothing techniques (e.g., moving
averages, exponential smoothing functions).
b. Seasonal analysis by calculating seasonal indices or using Excel’s seasonal decomposition tools.
c. Forecasting future electricity consumption based on historical data using Excel’s forecasting
functions, forecast sheet, or regression analysis.
d. For any prediction made, it is crucial to calculate evaluation metrics for time series. These may
include Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error
(RMSE), or Mean Absolute Percentage Error (MAPE).
e. Interpret the evaluation results obtained from the above step. A lower value for MAE, MSE,
RMSE, or MAPE indicates better predictive performance, as it signifies smaller prediction errors.
You should compare the evaluation metrics across different predictive models (e.g., moving
average, regression-based models, others) to determine which model performs best in forecasting
electricity consumption.
f. Create visualizations in Excel to visualize the prediction accuracy of your models. This could
include line charts comparing predicted values against actual values over time, with a focus on
how closely the predicted values align with the actual observations.
5. Project Report and Presentation:
a. Document your Excel analysis process and evaluation as clear as possible, including data
preprocessing, descriptive analytics, and predictive modeling techniques used.
b. Summarize findings, insights, and recommendations in a formal project report.
2
c. Each group or individual will present their project in class. Organize a presentation session
where you can present their Excel analyses, discuss methodology, and share insights with the
class. Max of 15-min presentation is required.
d. Reflect on the analysis and discuss key findings regarding electricity usage changes during the
COVID-19 pandemic.
e. Main take away from this project.
6. Final submission includes:
a. Excel file(s)
b. Write-up report in PDF named as report.pdf
c. PowerPoint presentation
7. Grading: the group project will be graded on the following component parts. If the project is done by a
group, all member of a group will receive the same grade. Total: 100 points.
a. Data Cleansing (10 points)
• Completeness of data cleansing tasks (removing/filling missing values, handling outliers,
ensuring data consistency).
• Effectiveness of data preprocessing techniques applied.
b. Descriptive Analytics (15 points)
• Accuracy of summary statistics calculations (mean, median, standard deviation, min, max).
• Comparison of summary statistics before and during the COVID-19 pandemic.
• Quality and relevance of visualizations (line charts, bar graphs) showcasing electricity usage
trends and patterns.
c. Predictive Analytics (30 points)
• Correct application of time series trend analysis techniques (regression, exponential
smoothing).
• Accuracy of seasonal analysis (calculating seasonal indices, using seasonal decomposition
tools).
• Effectiveness of forecasting future electricity consumption using Excel’s functions or
regression analysis.
• Calculation and interpretation of evaluation metrics (MAE, MSE, RMSE, MAPE) for
predictive models.
• Comparison of evaluation metrics across different predictive models to determine bestperforming model.
• Quality of visualizations depicting prediction accuracy and alignment with actual
observations.
d. Project Report and Presentation (30 points)
• Clarity and completeness of Excel analysis process documentation.
• Thoroughness of summarizing findings, insights, and recommendations in the report.
• Quality of PowerPoint presentation, including methodology discussion and insights shared
with the class.
3
• Reflection on analysis and discussion of key findings related to COVID-19 pandemic impact
on electricity usage.
• Clear articulation of main takeaways from the project.
e. Final Submission (10 points)
• Inclusion of Excel file(s) containing analysis and predictions.
• Submission of a well-written PDF report named “report.pdf” summarizing the analysis and
evaluation.
• Submission of a PowerPoint presentation for the project presentation session.
f. Bonus Points (5 points)
• Creativity and originality in data analysis techniques or visualizations.
• In-depth exploration of external factors impacting electricity consumption changes during the
pandemic.
• Effective communication of insights and recommendations in the presentation and report.
Order from us for quality, customized work in due time of your choice.