|
所在平台: Coursera |
课程主页: https://www.coursera.org/learn/excel-business-forecasting-regression
课程评论：没有评论
课程名称：Excel回归模型用于商业预测 课程概述：本课程旨在帮助学习者探索回归模型，以便利用这些模型进行商业预测。与时间序列模型不同，回归模型是因果模型，能够识别在商业中影响其他变量的特定变量。通过对这种因果关系的建模，我们能够进行预测，并为企业的需求进行规划。课程将涉及简单回归模型、多元回归模型、虚拟变量回归、季节变量回归以及自回归模型等不同形式的回归模型，针对特定商业场景进行预测，为组织生成商业智能。 课程大纲： 1. 欢迎与关键信息 - 介绍课程的重要信息。 2. 回归模型 - 探讨商业预测的背景和目的，了解使用回归模型的三种商业预测类型。学习回归模型的理论基础，理解解释变量和因变量之间的关系，重点关注单变量或简单回归，并学习如何使用回归诊断工具对模型进行批判性评估，然后根据组织的需求使用模型进行预测。 3. 多变量回归 - 将简单回归模型扩展为多个解释变量，进一步加深回归模型的理论基础，涉及多个因变量。学习如何使用回归诊断工具对多元回归模型进行批判性评估，并根据组织的需求进行预测。 4. 虚拟变量回归 - 将多元回归模型扩展为包括定性二进制解释变量，创建虚拟变量以处理二进制定性数据。学习如何对虚拟变量回归模型进行评估，并在组织需求框架下进行预测。 5. 季节虚拟回归 - 扩展二进制虚拟变量回归模型，纳入季节性变量，通过创建一系列虚拟变量来捕捉季节性。学习如何对季节虚拟回归模型进行评价，并为组织进行预测。本模块还将探讨自回归模型：其理论基础、创建方法、评价及其在商业预测中的应用。最后，学习如何将两个预测模型组合，形成复合预测。 此课程将为您提供掌握回归模型并利用其进行有效商业预测的必要技能。
Name:Welcome and Critical Information
Description:
Name:Regression Models
Description:In this module, we explore the context and purpose of business forecasting and the three types of business forecasting using regression models. We will learn the theoretical underpinning for a regression model, and understand the relationship between explanatory variables and dependent variables. We will first focus on single variable or simple regression, and learn how to critically evaluate the model using regression diagnostic tools and then use our models for forecasting to suit our organisation's needs.
Name:Multiple Variable Regression
Description:In this module, we extend the simple regression model to take in multiple explanatory variables. We will extend the theoretical underpinning for a regression model by involving multiple dependent variables. We will learn how to critically evaluate the multiple regression models using regression diagnostic tools and then use our models for forecasting to suit our organisation's needs.
Name:Dummy Variable Regression
Description:In this module, we extend the multiple regression model to take in qualitative binary explanatory variables. We will extend the theoretical underpinning for a multiple regression model by creating dummy variables for binary qualitative data. We will learn how to critically evaluate the dummy variable regression models using regression diagnostic tools and then use our models for forecasting to suit our organisation's needs.
Name:Seasonal Dummy Regression
Description:In this module, we extend the binary dummary variable regression model to take in seasonal variables. We will extend the theoretical underpinning for a binary dummy variable regression model by creating a series of dummy variables to capture seasonality. We will learn how to critically evaluate the seasonal dummy regression models using regression diagnostic tools and then use our models for forecasting to suit our organisation's needs. In this module we will also explore autoregressions - their theoretical underpinning, creating an autoregression, critically evaluating this, and utilising our model for business forecasting. We will end the module by learning how to create a composite forecast by combining two forecasts across this course and the first course in this specialisation.
This course allows learners to explore Regression Models in order to utilise these models for business forecasting. Unlike Time Series Models, Regression Models are causal models, where we identify certain variables in our business that influence other variables. Regressions model this causality, and then we can use these models in order to forecast, and then plan for our business' needs. We will explore simple regression models, multiple regression models, dummy variable regressions, seasonal variable regressions, as well as autoregressions. Each of these are different forms of regression models, tailored to unique business scenarios, in order to forecast and generate business intelligence for organisations.
