Data Science Intermediate
3
Days
Linear Regression
Day 1
Get In Touch
PT. Algoritma Data Indonesia.
RDTX Square, 9th Floor. Setiabudi, DKI Jakarta 12930.
WhatsApp: 0816-692-471
Email: community@algorit.ma
Course details :
This intermediate-level workshop is designed to help you master regression models and a closely related concept, generalized linear models. Regression models has been called the “workhorse” of data science, and often is the first tool of choice for any machine learning task. Students will learn to fit, interpret and examine simple and multiple linear regression models as we build regression models modeled after real-life business cases.
The course will take the student beyond fitting a regression line and interpreting our regression coefficients. We’ll discuss concepts such as confounding variables, and using various diagnostic tools to evaluate our models. Student will also replicate notable case studies and observe how linear methods are equally, if not more, successful at performing complex machine learning tasks than many of the newer AI technologies.
Data is the new oil? No: Data is the new soil. ~ David McCandless
Please bring along:
- 1x Laptop
- Purchased ticket (from organizer’s website)
In a nutshell
Date to be advised
Trainer
Samuel Chan
samuel@algorit.ma
Detailed Syllabus
Syllabus: Data Science Intermediate (I)
Day 1
Regression Models
- Why is linear models so often the first tool of choice for statisticians and data scientists
- Linear models, visually
- Estimating the regression coefficients
- Regression through the origin
Residual Variation and Model Interpretation
- Properties of the residuals
- Estimating our error term
- Total variability vs Regression variability
- t-Statistics and P-value
- Confidence level and Confidence Intervals
Day 2
Least Squares Estimates in Multivariable Regressions
- Regression with two regressors
- Simulation: Multivariable regressions
- Estimating the slope: 3 methods
- Putting it together
Key Concepts in Regression
- Adjustment
- Model diagnostics
- Model selection
- ANOVA
- Tips on picking and fitting a regression model
Day 3
Logistic Regression and Poisson Regressions
- Generalized linear model
- The logit() and inverted logit() function
- A closer look at the sigmoid curve
- Logistic regression for a binary response variable
- Poisson regression examples
Extra tips and techniques
- Applying regression analysis in the context of machine learning
- Polynomial regression
- Tools to evaluate model fit
- Techniques to test for outliers, their leverage and influence
Artificial Neural Network
- Cost Function
- Backpropagational and Feed-Forward algorithms
- Implementing ANN for optical character recognition
- Fine-tuning our Neural Nets parameter
This workshop will cost 3 workshop credits for subscribers. Non-subscribers are welcomed to participate at a cost of IDR3,000,000.
Workshop Receivables:
Data Science Intermediate Series
Workshops in our Data Science Intermediate series are tailored to R programmers that are taking their first steps into data science and machine learning.
Students are assumed to have a working knowledge of R and ideally some proficiency in statistics / mathematics / algebra. Consider taking our Data Science Fundamentals workshops instead if you have no prior programming experience or statistics knowledge:
Students work through tons of real-life examples using sample datasets donated by our team of mentors and corporate partners. We believe in a learn-by-building approach, and we employ instructors who are uncompromisingly passionate about your growth and education.
