Bis551 Business Intelligence And Data Mining Case 2 Submissi

Bis551 Business Intelligence And Data Miningcase 2to Be Submitted In W

List and explain the three major categories of business reports. What are the most important assumptions in linear regression? Explain by example Describe categorical and nominal data. Give example for each A common way of introducing data warehousing is to refer to its fundamental characteristics. Describe three characteristics of data warehousing. What are the four processes that define a closed-loop BPM cycle? Give example for each

Paper For Above instruction

Business reports are essential tools that facilitate decision-making by providing summaries, insights, and analyses of business data. These reports are generally categorized into three major types: operational reports, managerial reports, and strategic reports. Each serves distinct purposes and audiences within an organization.

Operational reports are designed for day-to-day management. They provide real-time or periodic data that help monitor current performance and facilitate immediate decision-making. Examples include sales reports, inventory reports, and daily transaction summaries. These reports are typically detailed and focus on operational efficiency.

Managerial reports support middle management in tactical decision-making. They often include summarized data, trends, and comparisons to assist managers in resource allocation, performance evaluations, and short-term planning. An example would be monthly sales performance reports, which may compare regional sales figures over a period to identify areas needing attention.

Strategic reports are aimed at senior management and focus on long-term planning and strategic initiatives. They analyze market trends, forecasts, and competitive positioning. Examples include industry analysis reports and corporate growth projections. These reports often incorporate graphical representations for better visualization of complex data.

The assumptions in linear regression are critical for ensuring the validity of the model. The primary assumptions include linearity, independence, homoscedasticity, normality, and no multicollinearity.

Linearity assumes a straight-line relationship between the independent variables and the dependent variable. For example, suppose we want to predict sales based on advertising expenditure; a linear relationship suggests that increasing advertising directly correlates with higher sales.

Independence stipulates that residuals are independent, meaning that the errors associated with one observation are not influenced by errors in others. This is often assessed through study design and residual plots.

Homoscedasticity means that the variance of residuals is constant across all levels of the independent variables. For instance, in predicting house prices, the spread of errors should be similar regardless of whether houses are inexpensive or expensive.

Normality assumes that the residuals are approximately normally distributed. This can be evaluated via histograms or Q-Q plots. For example, in a regression analyzing customer satisfaction, the residuals should form a bell-shaped curve.

Multicollinearity occurs when independent variables are highly correlated, which can distort the estimates of coefficients. For example, using both total income and disposable income as predictors in a regression could introduce multicollinearity because they are related.

Categorical data refers to variables that contain label values rather than numeric measures, often representing classifications or groups. Nominal data is a type of categorical data where the categories have no intrinsic order.

An example of categorical data is "Marital Status" with categories such as single, married, divorced, and widowed. Each category labels a different group but has no ordinal relationship.

Nominal data could also include "Color" with categories like red, blue, green, and yellow. Since these categories do not have a natural order, they are nominal rather than ordinal.

Data warehousing is a core component of business intelligence systems, enabling organizations to consolidate and analyze vast amounts of data from various sources. Its fundamental characteristics include subject orientation, integrated data, and non-volatile storage.

Subject orientation refers to organizing data around key subjects such as customers, products, or sales rather than around applications or transactions. For instance, a data warehouse may have separate schemas for customer data and sales data, facilitating analysis of customer behavior and sales performance.

Integration involves consolidating data from different sources into a coherent store, resolving conflicts in formats, naming conventions, and data quality issues. An example is combining transaction data from multiple regional offices into a unified sales database.

Non-volatility signifies that once data is entered into the warehouse, it is stable and not updated randomly. Instead, it is periodically refreshed, enabling historical analysis. For example, a data warehouse might keep monthly summaries of sales data over several years for trend analysis.

A closed-loop Business Process Management (BPM) cycle involves four processes: design, model, execute, and monitor & optimize. These processes form a continuous feedback loop that enhances process efficiency.

Design involves defining and modeling the process, such as mapping out the steps involved in order processing in an e-commerce platform.

Model refers to creating detailed representations of the process, including workflows, decision points, and resource allocations. For example, designing a process flowchart for customer onboarding.

Execute involves implementing the modeled process within IT systems or operational environments. An example is deploying the order processing workflow into a supply chain management system.

Monitor & optimize entail tracking the process performance using key performance indicators (KPIs) and making adjustments for improved efficiency. For instance, analyzing order fulfillment times and implementing changes to reduce delays.

References

• Inmon, W. H. (2005). Building the Data Warehouse. John Wiley & Sons.
• Kimball, R., Ross, M. (2013). The Data Warehouse Toolkit: The Definitive Guide to Dimensional Modeling. John Wiley & Sons.
• Montgomery, D. C. (2017). Design and Analysis of Experiments. John Wiley & Sons.
• Gelman, A., Hill, J. (2006). Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press.
• Rosenberg, M. (2004). Data Mining Techniques: For Marketing, Sales, and Customer Relationship Management. Wiley.
• Louviere, J. J., Hensher, D. A., Swait, J. D. (2000). Stated Choice Methods: Analysis and Applications. Cambridge University Press.
• Delen, D., Demirkan, H. (2013). Data, information, and analytics as services. Decision Support Systems, 55(1), 359-363.
• Pei, J., Han, J., Kamber, M. (2014). Data Mining: Concepts and Techniques. Morgan Kaufmann.
• Shmueli, G., Bruce, P. C., Gedeck, P. (2020). Data Mining for Business Analytics. Pearson.
• Turban, E., Sharda, R., Dursun, P., et al. (2018). Business Intelligence and Analytics: Systems for Decision Support. Pearson.