Predictive and visual analytics for strategic business planning

Quarlo Solutions需要对战略业务规划的预测和可视化分析进行研究。一个特别值得关注的问题是对加拿大经济，特别是加拿大信用社的业务进行预测。该研究项目的主要目标是（1）改善加拿大信用社业务的预测模型，该模型结合了有关全球，国家和地方经济因素的公开和私人经济数据，以及（2）改善以易于理解的格式显示信用社健康和绩效信息的仪表板。给定一个基于自回归综合移动平均（ARIMA）的初始模型，将评估各种预测分析技术，包括向量自回归（VAR）、机器学习（分类学习和深度学习）和数据挖掘，以确定它们是否能提高预测准确性。仪表板将得到改进，以显示各种数据源的输入和汇总输出。将进行研究，以确定图形用户界面（GUI）的适当方式，在实际经济和业务表现严重偏离预测的情况下显示警报。这一信息将提供给各个信用合作社，并在汇总价值的范围内提出。在研究的评估部分，首先将改进的预测模型进行实施，并使用一些人工数据进行验证。接下来，我们将把改进后的模型应用到历史数据中，以确定其有效性。将模拟各种条件，以确保生成警报。另外，作为评估的一部分，我们将在仪表板上进行详细的软件测试，以确保所有必需的功能都存在，并正确工作和可靠性。最后，将在执行下一年战略业务规划的背景下，与联合国开发计划署人员一起对软件原型进行评估。

Celero Solutions requires research on predictive and visual analytics for strategic business planning. A particular concern is making predictions about the Canadian economy and especially the business of Canadian credit unions. The main objectives of the research project are (1) to improve a predictive model of the Canadian credit union business that incorporates publicly available and private economic data concerning global, national, and local economic factors, and (2) to improve a dashboard that displays information about credit union health and performance in an easy to understand format. Given an initial model based on auto-regressive integrated moving averages (ARIMA), a wide variety of predictive analytics techniques including vector auto-regression (VAR), machine learning (classification learning and deep learning), and data mining will be evaluated to determine whether they improve predictive accuracy. The dashboard will be improved to display the input and summarized output for a large variety of data sources. Research will be performed to determine an appropriate way for the graphical user interface (GUI) to display alerts for cases where the actual economic and business performance significantly deviates from forecasts. This information will be given for individual credit unions and also presented in the context of aggregated values. In the evaluation part of the research, first the improved predictive model will be implemented and verified using some artificial data. Next, we will apply the improved model to historical data to determine its effectiveness. A variety of conditions will be simulated to ensure that alerts are produced. Separately, as part of the evaluation, we will perform detailed software testing on the dashboard to ensure that all required functionality is present and works correctly and reliability. Finally, the software prototype will be evaluated in association with Celero personnel in the context of performing strategic business planning for the next year.