Fundamentals

Imagine a small bakery, aromas of fresh bread mingling with the scent of brewing coffee, yet behind the charming facade lies a daily scramble to predict just how many loaves to bake, how much flour to order, and how many staff to schedule ● this very scenario underscores the often-unseen data deluge even the smallest businesses navigate daily.

Decoding Predictive Business Analytics For Small Businesses

Predictive business analytics, often perceived as the domain of corporate giants with sprawling data centers, holds surprisingly tangible benefits for small to medium-sized businesses (SMBs). At its core, predictive analytics Meaning ● Strategic foresight through data for SMB success. utilizes historical data, statistical algorithms, and machine learning Meaning ● Machine Learning (ML), in the context of Small and Medium-sized Businesses (SMBs), represents a suite of algorithms that enable computer systems to learn from data without explicit programming, driving automation and enhancing decision-making. techniques to forecast future outcomes. Think of it less as gazing into a crystal ball and more as using sophisticated tools to analyze past trends to anticipate what might come next, providing SMBs with a crucial edge in an increasingly competitive landscape.

Why Should SMBs Care About Prediction?

For SMBs operating with tight margins and limited resources, the ability to foresee future trends translates directly into smarter decisions and reduced risks. Consider inventory management ● overstocking ties up precious capital in unsold goods, while understocking leads to lost sales and dissatisfied customers. Predictive analytics offers a pathway to optimize inventory levels by forecasting demand with greater accuracy, ensuring the bakery, for instance, bakes just the right amount of sourdough each morning. This precision extends beyond inventory, touching every facet of an SMB’s operations, from marketing campaigns to staffing decisions.

Practical Applications In Everyday SMB Operations

The beauty of predictive analytics for SMBs Meaning ● Predictive Analytics for SMBs: Using data to foresee trends and make smarter decisions for growth and efficiency. lies in its practical, down-to-earth applications. It is not about abstract theories; it is about tangible improvements in day-to-day operations. Let’s explore some key areas where predictive analytics can make a real difference:

Sales Forecasting ● Anticipating Customer Demand

Predicting sales accurately is paramount for any SMB. Traditional methods often rely on gut feeling or simple year-over-year comparisons, which can be unreliable and miss crucial nuances. Predictive analytics, however, can analyze historical sales data, seasonal trends, marketing campaign performance, and even external factors like local events or weather patterns to generate more precise sales forecasts. This allows SMBs to proactively adjust inventory, staffing, and marketing strategies to meet anticipated demand, minimizing waste and maximizing revenue.

Customer Behavior Analysis ● Understanding Your Customers

Understanding customer behavior Meaning ● Customer Behavior, within the sphere of Small and Medium-sized Businesses (SMBs), refers to the study and analysis of how customers decide to buy, use, and dispose of goods, services, ideas, or experiences, particularly as it relates to SMB growth strategies. is no longer a luxury; it is a necessity. Predictive analytics can sift through customer data Meaning ● Customer Data, in the sphere of SMB growth, automation, and implementation, represents the total collection of information pertaining to a business's customers; it is gathered, structured, and leveraged to gain deeper insights into customer behavior, preferences, and needs to inform strategic business decisions. ● purchase history, website interactions, survey responses ● to identify patterns and predict future behavior. This enables SMBs to personalize marketing efforts, anticipate customer needs, and improve customer retention. For example, a small online retailer can use predictive analytics to identify customers likely to churn and proactively offer them personalized discounts or loyalty rewards, fostering stronger customer relationships.

Inventory Optimization ● Balancing Supply and Demand

Efficient inventory management is crucial for SMB profitability. Holding too much inventory incurs storage costs and increases the risk of obsolescence, while insufficient inventory leads to stockouts and lost sales. Predictive analytics can optimize inventory levels by forecasting demand fluctuations, considering lead times, and identifying optimal reorder points. This ensures SMBs maintain adequate stock levels to meet customer demand without tying up excessive capital in inventory, improving cash flow and operational efficiency.

Marketing Campaign Effectiveness ● Reaching the Right Audience

Marketing budgets for SMBs are often limited, making it imperative to maximize the return on every marketing dollar spent. Predictive analytics can enhance marketing campaign effectiveness by identifying the most receptive customer segments, predicting campaign response rates, and optimizing channel selection. A local gym, for example, can use predictive analytics to target marketing efforts towards individuals most likely to be interested in a membership based on demographic data, past engagement, and fitness trends, leading to higher conversion rates and reduced marketing costs.

Risk Management ● Mitigating Potential Threats

SMBs face various risks, from financial instability to operational disruptions. Predictive analytics can play a crucial role in risk management by identifying potential risks early on and enabling proactive mitigation strategies. For instance, a small lending institution can use predictive analytics to assess credit risk more accurately, reducing loan defaults. Similarly, a manufacturing SMB can use predictive maintenance to foresee equipment failures, minimizing downtime and production disruptions.

Predictive analytics empowers SMBs to transition from reactive problem-solving to proactive opportunity creation.

Getting Started ● Practical Steps For SMBs

Implementing predictive analytics might seem daunting, but SMBs can start with manageable steps and gradually scale their efforts. Here are some practical starting points:

Identify Key Business Questions

Begin by pinpointing specific business questions that predictive analytics can help answer. Instead of broadly aiming to “improve business,” focus on concrete questions like ● “How can we reduce customer churn?”, “What products will be in high demand next quarter?”, or “How can we optimize our staffing levels during peak hours?”. Clearly defined questions will guide data collection and analysis efforts, ensuring focused and impactful results.

Leverage Existing Data

SMBs often underestimate the wealth of data they already possess. Transaction records, website analytics, customer relationship management (CRM) systems, social media data ● these are all potential sources of valuable insights. Start by auditing existing data sources and identifying data relevant to the business questions identified earlier. Focus on utilizing readily available data before investing in extensive data collection infrastructure.

Choose the Right Tools and Technologies

The landscape of analytics tools is vast, but SMBs do not need to invest in complex, enterprise-grade solutions initially. Many user-friendly, affordable tools are specifically designed for SMBs. Cloud-based analytics platforms, spreadsheet software with advanced statistical functions, and even specialized SMB-focused analytics applications can provide powerful predictive capabilities without requiring extensive technical expertise or significant upfront investment. Selecting tools that align with the SMB’s technical capabilities and budget is crucial for successful implementation.

Focus On Actionable Insights

The ultimate goal of predictive analytics is to generate actionable insights that drive tangible business improvements. Avoid getting bogged down in complex analyses that do not translate into practical actions. Focus on extracting insights that can inform decision-making and lead to measurable outcomes.

For example, a sales forecast is only valuable if it prompts adjustments to inventory levels or marketing strategies. Prioritize insights that are clear, concise, and directly applicable to SMB operations.

Start Small and Iterate

Implementing predictive analytics is an iterative process. Start with a pilot project in a specific area of the business, such as sales forecasting Meaning ● Sales Forecasting, within the SMB landscape, is the art and science of predicting future sales revenue, essential for informed decision-making and strategic planning. or customer churn Meaning ● Customer Churn, also known as attrition, represents the proportion of customers that cease doing business with a company over a specified period. prediction. Begin with simple models and gradually refine them as you gain experience and collect more data.

Regularly evaluate the results, learn from both successes and failures, and iterate to improve the accuracy and effectiveness of predictive analytics efforts. A phased approach allows SMBs to learn, adapt, and build confidence in predictive analytics capabilities over time.

Predictive analytics is not an exclusive privilege reserved for large corporations. SMBs, regardless of size or industry, can harness its power to gain a competitive edge. By embracing a data-driven mindset and taking incremental steps, SMBs can unlock the predictive potential within their existing data, making smarter decisions, optimizing operations, and paving the way for sustainable growth.

The journey into predictive analytics for SMBs begins not with complex algorithms, but with a simple question ● “What future outcomes do we need to anticipate to make better decisions today?”. Answering this question is the first step towards transforming data from a passive record of the past into a powerful predictor of future success.

Below is a table showcasing potential applications of predictive analytics across different SMB functions:

And here is a list of initial steps SMBs can take to begin their predictive analytics journey:

1. Identify Key Business Questions ● Determine specific areas where prediction can add value.
2. Audit Existing Data ● Assess available data sources and their relevance.
3. Choose User-Friendly Tools ● Select affordable and accessible analytics platforms.
4. Focus on Actionable Insights ● Prioritize insights that drive practical decisions.
5. Start Small and Iterate ● Implement pilot projects and gradually expand.

The path to predictive analytics adoption for SMBs is not a sprint, but a marathon of continuous learning and refinement. By taking these fundamental steps, SMBs can begin to harness the predictive power of their data, transforming uncertainty into informed action and laying a solid foundation for future growth and resilience.

Intermediate

Consider this ● a recent study indicated that SMBs adopting data-driven decision-making processes witnessed a 23% increase in customer acquisition and a 19% boost in profitability compared to their less data-savvy counterparts. These figures are not mere statistics; they represent a tangible shift in the SMB competitive landscape, where predictive business analytics Meaning ● Business Analytics for SMBs: Smart decision-making using data to drive growth and efficiency. is rapidly transitioning from a ‘nice-to-have’ to a ‘must-have’ capability.

Deep Dive Into Predictive Modeling Techniques

Moving beyond the basic understanding, SMBs ready to delve deeper into predictive analytics need to familiarize themselves with core modeling techniques. These techniques, while rooted in statistical and machine learning principles, are increasingly accessible through user-friendly platforms and services. Understanding the nuances of these models allows SMBs to select the most appropriate approach for their specific business challenges and data characteristics.

Regression Analysis ● Unveiling Relationships Between Variables

Regression analysis is a foundational technique used to model the relationship between a dependent variable (the outcome we want to predict, such as sales revenue) and one or more independent variables (factors that might influence the outcome, such as marketing spend or seasonality). Linear regression, a common type, assumes a linear relationship, while more complex forms like polynomial or logistic regression can capture non-linear patterns. For an SMB, regression analysis Meaning ● Regression Analysis, a statistical methodology vital for SMBs, facilitates the understanding of relationships between variables to predict outcomes. can be used to predict sales based on advertising expenditure, forecast customer lifetime value Meaning ● Customer Lifetime Value (CLTV) for SMBs is the projected net profit from a customer relationship, guiding strategic decisions for sustainable growth. based on engagement metrics, or estimate inventory needs based on historical demand and promotional activities.

Time Series Analysis ● Forecasting Trends Over Time

Time series analysis is specifically designed for data collected over time, such as daily sales figures, website traffic, or stock prices. Techniques like ARIMA (Autoregressive Integrated Moving Average) and Exponential Smoothing can identify patterns, seasonality, and trends in time series data to forecast future values. SMBs can leverage time series analysis Meaning ● Time Series Analysis for SMBs: Understanding business rhythms to predict trends and make data-driven decisions for growth. for sales forecasting, demand planning, and predicting website traffic fluctuations, enabling proactive resource allocation and operational adjustments.

Classification Models ● Categorizing and Predicting Group Membership

Classification models are used to predict categorical outcomes, such as whether a customer will churn (yes/no), whether a transaction is fraudulent (fraudulent/not fraudulent), or which customer segment a new customer belongs to (segment A, B, or C). Algorithms like decision trees, support vector machines (SVMs), and Naive Bayes are commonly used for classification tasks. SMBs can apply classification models to predict customer churn, identify high-potential leads, segment customers for targeted marketing, and detect fraudulent transactions, enhancing customer retention, marketing effectiveness, and risk management.

Clustering Analysis ● Discovering Hidden Patterns in Data

Clustering analysis is an unsupervised learning technique that groups similar data points together based on their characteristics without predefined categories. Algorithms like k-means clustering and hierarchical clustering can identify natural groupings in customer data, product data, or transaction data. SMBs can utilize clustering to segment customers based on purchasing behavior, identify product categories that are frequently bought together, or discover geographic clusters of high-demand areas, informing targeted marketing, product recommendations, and location-based strategies.

The selection of the right predictive model hinges on a clear understanding of the business problem and the nature of available data.

Data Preprocessing and Feature Engineering ● Preparing Data For Prediction

The accuracy of predictive models Meaning ● Predictive Models, in the context of SMB growth, refer to analytical tools that forecast future outcomes based on historical data, enabling informed decision-making. heavily relies on the quality and preparation of the input data. Data preprocessing and feature engineering are crucial steps in transforming raw data into a format suitable for modeling. These processes involve cleaning data, handling missing values, transforming variables, and creating new features that enhance the predictive power of the models.

Data Cleaning and Handling Missing Values

Real-world data is often messy, containing errors, inconsistencies, and missing values. Data cleaning involves identifying and correcting errors, resolving inconsistencies, and handling missing values. Techniques for handling missing values include imputation (replacing missing values with estimated values) or removal of records with missing values. Clean and consistent data is essential for building robust and reliable predictive models.

Feature Scaling and Transformation

Feature scaling involves normalizing or standardizing numerical features to ensure that features with larger scales do not disproportionately influence the models. Feature transformation involves applying mathematical transformations to variables to improve their distribution or capture non-linear relationships. For example, logarithmic transformation can be applied to skewed data, and polynomial features can be created to capture non-linear relationships between variables. Appropriate feature scaling and transformation can significantly improve model performance.

Feature Selection and Engineering

Feature selection involves selecting the most relevant features from the dataset to reduce dimensionality, improve model interpretability, and prevent overfitting. Feature engineering involves creating new features from existing ones that might capture additional information or improve model accuracy. For instance, creating a “customer engagement score” feature from website visit frequency, time spent on site, and pages viewed can provide a more informative feature than individual metrics. Effective feature selection and engineering are critical for building parsimonious and accurate predictive models.

Evaluating Predictive Model Performance ● Ensuring Accuracy and Reliability

Building a predictive model is only half the battle; evaluating its performance is equally crucial to ensure its accuracy and reliability. Various metrics and techniques are used to assess model performance, depending on the type of model and the business objective. Understanding these evaluation methods allows SMBs to objectively assess model effectiveness and make informed decisions about model deployment and refinement.

Accuracy, Precision, Recall, and F1-Score

For classification models, metrics like accuracy (overall correctness), precision (proportion of correctly predicted positives out of all predicted positives), recall (proportion of correctly predicted positives out of all actual positives), and F1-score (harmonic mean of precision and recall) are commonly used. These metrics provide a comprehensive assessment of model performance in classification tasks, particularly when dealing with imbalanced datasets where one class is significantly more frequent than the other.

RMSE, MAE, and R-Squared

For regression models, metrics like Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R-squared (coefficient of determination) are used to evaluate prediction accuracy. RMSE and MAE measure the average magnitude of errors between predicted and actual values, while R-squared measures the proportion of variance in the dependent variable explained by the model. These metrics help assess the goodness-of-fit and predictive accuracy of regression models.

Cross-Validation and Holdout Sets

Cross-validation techniques, such as k-fold cross-validation, are used to assess model generalization performance and prevent overfitting. Holdout sets, where a portion of the data is held back for final model evaluation, provide an independent assessment of model performance on unseen data. These techniques ensure that model evaluation is robust and provides a realistic estimate of model performance in real-world applications.

Here is a table summarizing common predictive modeling Meaning ● Predictive Modeling empowers SMBs to anticipate future trends, optimize resources, and gain a competitive edge through data-driven foresight. techniques and their applications for SMBs:

And here is a list of key considerations for data preprocessing and model evaluation:

• Data Quality is Paramount ● Clean and consistent data is crucial for accurate predictions.
• Feature Engineering Enhances Models ● Creating relevant features improves predictive power.
• Model Evaluation is Essential ● Objectively assess model performance using appropriate metrics.
• Cross-Validation Prevents Overfitting ● Ensure models generalize well to unseen data.
• Iterative Refinement is Key ● Continuously improve models based on evaluation results.

As SMBs progress on their predictive analytics journey, a deeper understanding of modeling techniques, data preprocessing, and model evaluation becomes indispensable. This intermediate level of knowledge empowers SMBs to build more sophisticated and accurate predictive models, extract richer insights from their data, and unlock a greater spectrum of business benefits. The transition from basic applications to more advanced modeling marks a significant step towards data-driven maturity, enabling SMBs to leverage predictive analytics for strategic advantage and sustained growth.

Advanced

Consider the trajectory of market leaders ● businesses that not only react to change but proactively shape it often leverage predictive analytics as a strategic weapon, achieving a competitive advantage that extends beyond operational efficiency to encompass market anticipation and innovation. For SMBs aspiring to disrupt markets and redefine industry norms, advanced predictive analytics is not merely a tool for optimization; it is the engine of strategic foresight.

Integrating Advanced Machine Learning and AI Techniques

At the advanced level, SMBs can explore the integration of sophisticated machine learning (ML) and artificial intelligence (AI) techniques to unlock predictive capabilities that go beyond traditional statistical methods. These advanced techniques, fueled by increasing computational power and cloud-based accessibility, offer the potential to model complex, non-linear relationships, handle massive datasets, and automate predictive processes at scale.

Deep Learning ● Unlocking Complex Patterns in High-Dimensional Data

Deep learning, a subfield of machine learning inspired by the structure of the human brain, utilizes artificial neural networks with multiple layers (deep neural networks) to learn intricate patterns from vast amounts of data. Deep learning excels in handling high-dimensional data, such as images, text, and audio, and can automatically extract relevant features, reducing the need for manual feature engineering. For SMBs, deep learning can be applied to analyze customer sentiment Meaning ● Customer sentiment, within the context of Small and Medium-sized Businesses (SMBs), Growth, Automation, and Implementation, reflects the aggregate of customer opinions and feelings about a company’s products, services, or brand. from social media text, predict customer behavior from complex website interaction patterns, or optimize visual merchandising based on image analysis of customer traffic flow in retail stores.

Ensemble Methods ● Combining Multiple Models for Enhanced Accuracy

Ensemble methods combine predictions from multiple individual models to create a more robust and accurate predictive model. Techniques like Random Forests, Gradient Boosting Machines (GBM), and XGBoost (Extreme Gradient Boosting) build ensembles of decision trees, leveraging the wisdom of crowds to improve prediction accuracy and reduce overfitting. SMBs can employ ensemble methods to enhance the accuracy of sales forecasts, customer churn predictions, and risk assessments, achieving superior predictive performance compared to single models.

Natural Language Processing (NLP) ● Extracting Insights from Text Data

Natural Language Processing (NLP) enables computers to understand, interpret, and generate human language. NLP techniques can analyze text data from customer reviews, social media posts, surveys, and customer service Meaning ● Customer service, within the context of SMB growth, involves providing assistance and support to customers before, during, and after a purchase, a vital function for business survival. interactions to extract valuable insights about customer sentiment, preferences, and emerging trends. SMBs can leverage NLP to monitor brand reputation, identify customer pain points, personalize customer communication, and automate customer service responses, enhancing customer understanding and engagement.

Reinforcement Learning ● Optimizing Decisions Through Trial and Error

Reinforcement learning is a type of machine learning where an agent learns to make optimal decisions in an environment through trial and error, receiving rewards or penalties for its actions. Reinforcement learning is particularly well-suited for dynamic and complex decision-making problems, such as pricing optimization, inventory management in fluctuating markets, and personalized recommendation systems. SMBs can explore reinforcement learning to optimize pricing strategies in real-time, personalize product recommendations based on individual customer interactions, and automate complex operational decisions in dynamic environments.

Advanced ML and AI techniques empower SMBs to move beyond descriptive analytics to truly prescriptive and autonomous predictive capabilities.

Data Governance and Ethical Considerations in Advanced Analytics

As SMBs embrace advanced predictive analytics, data governance Meaning ● Data Governance for SMBs strategically manages data to achieve business goals, foster innovation, and gain a competitive edge. and ethical considerations become paramount. Handling sensitive customer data, ensuring model fairness and transparency, and complying with data privacy Meaning ● Data privacy for SMBs is the responsible handling of personal data to build trust and enable sustainable business growth. regulations are critical aspects of responsible AI implementation. Establishing robust data governance frameworks and adhering to ethical principles are essential for building trust, mitigating risks, and ensuring the sustainable and ethical use of advanced analytics.

Data Privacy and Security

Advanced analytics often involves processing large volumes of sensitive customer data, making data privacy and security Meaning ● Data privacy, in the realm of SMB growth, refers to the establishment of policies and procedures protecting sensitive customer and company data from unauthorized access or misuse; this is not merely compliance, but building customer trust. a top priority. SMBs must implement robust data security measures to protect data from unauthorized access, breaches, and cyber threats. Complying with data privacy regulations, such as GDPR (General Data Protection Regulation) and CCPA (California Consumer Privacy Act), is essential for maintaining customer trust and avoiding legal repercussions. Data anonymization, encryption, and secure data storage practices are crucial components of data privacy and security in advanced analytics.

Model Fairness and Bias Mitigation

Predictive models can inadvertently perpetuate or amplify biases present in the training data, leading to unfair or discriminatory outcomes. Ensuring model fairness and mitigating bias is crucial for ethical AI implementation. SMBs should implement techniques to detect and mitigate bias in their models, such as fairness-aware algorithms, bias auditing, and diverse data collection practices. Transparency in model development and deployment processes is also essential for building trust and accountability.

Transparency and Explainability

Advanced machine learning models, particularly deep learning models, can be complex and opaque, making it challenging to understand why they make specific predictions (the “black box” problem). Transparency and explainability are increasingly important, especially in applications where decisions have significant impact on individuals or businesses. SMBs should explore techniques to improve model explainability, such as SHAP (SHapley Additive exPlanations) values and LIME (Local Interpretable Model-agnostic Explanations), to gain insights into model decision-making processes and build trust in AI-driven predictions.

Scaling Predictive Analytics for SMB Growth and Automation

The ultimate value of advanced predictive analytics for SMBs lies in its ability to drive growth and automation at scale. Integrating predictive insights Meaning ● Predictive Insights within the SMB realm represent the actionable intelligence derived from data analysis to forecast future business outcomes. into core business processes, automating predictive workflows, and leveraging cloud-based platforms are key strategies for scaling predictive analytics and maximizing its impact on SMB performance.

Integrating Predictive Insights into Business Processes

To realize the full potential of predictive analytics, SMBs must seamlessly integrate predictive insights into their core business processes. This involves embedding predictive models into operational systems, providing real-time predictive dashboards to decision-makers, and establishing feedback loops to continuously refine models based on real-world performance. Integrating predictive analytics into workflows transforms data-driven insights from isolated reports into actionable intelligence that guides day-to-day operations and strategic decision-making.

Automating Predictive Workflows

Automation is crucial for scaling predictive analytics and maximizing efficiency. SMBs can automate data preprocessing pipelines, model training and deployment processes, and predictive reporting workflows using cloud-based platforms and automation tools. Automating repetitive tasks frees up valuable time for data scientists and business analysts to focus on higher-value activities, such as model innovation, strategic analysis, and business problem-solving. Automation enables SMBs to deploy and maintain predictive analytics solutions at scale with limited resources.

Leveraging Cloud-Based Predictive Analytics Platforms

Cloud-based predictive analytics platforms provide SMBs with access to scalable computing resources, pre-built machine learning algorithms, and user-friendly interfaces, democratizing access to advanced analytics Meaning ● Advanced Analytics, in the realm of Small and Medium-sized Businesses (SMBs), signifies the utilization of sophisticated data analysis techniques beyond traditional Business Intelligence (BI). capabilities. Cloud platforms eliminate the need for significant upfront investments in infrastructure and expertise, allowing SMBs to rapidly deploy and scale predictive analytics solutions. Leveraging cloud platforms enables SMBs to focus on extracting business value from predictive analytics without being constrained by technical complexities or resource limitations.

Below is a table showcasing advanced predictive analytics techniques and their potential impact on SMBs:

And here is a list of key considerations for advanced predictive analytics implementation:

• Embrace Advanced Techniques ● Explore deep learning, ensemble methods, and NLP for complex problems.
• Prioritize Data Governance ● Establish robust data privacy, security, and ethical frameworks.
• Ensure Model Fairness and Transparency ● Mitigate bias and improve model explainability.
• Integrate Predictive Insights ● Embed analytics into core business processes for actionable intelligence.
• Leverage Cloud Platforms ● Utilize cloud resources for scalability, accessibility, and automation.

The journey to advanced predictive analytics is a strategic evolution for SMBs seeking to not just compete, but to lead. By embracing advanced ML and AI techniques, prioritizing data governance and ethics, and scaling predictive analytics through integration and automation, SMBs can unlock a new era of strategic foresight, operational agility, and sustainable growth. The future of SMB competitiveness is inextricably linked to the intelligent and ethical application of advanced predictive analytics, transforming data from a historical record into a compass guiding future success.

References

• Provost, Foster, and Tom Fawcett. Data Science for Business ● What You Need to Know About Data Mining and Data-Analytic Thinking. O’Reilly Media, 2013.
• Shmueli, Galit, Peter C. Bruce, Peter Gedeck, and Nitin R. Patel. Data Mining for Business Analytics ● Concepts, Techniques, and Applications in Python. Wiley, 2020.
• James, Gareth, Daniela Witten, Trevor Hastie, and Robert Tibshirani. An Introduction to Statistical Learning. Springer, 2013.