Top 10 Statistics Mistakes Made by Data Scientists

This video present the top 10 statistics mistakes made by data scientist in daily life by doing business tasks, building machine learning or deep learning models, coding statistical models and other fun stuff in real data science.

Following mistakes are explained in the video:
#1. Not fully understanding the objective function.
#2. Not having a hypothesis on why something should work.
#3. Not looking at the data before interpreting results.
#4. Not having a naive baseline model.
#5. Incorrect out-sample testing.
#6. Incorrect out-sample testing: applying preprocessing to full dataset.
#7. Incorrect out-sample testing: cross-sectional data & panel data.
#8. Not considering which data is available at point of decision.
#9. Subtle Overtraining.
#10. "need more data" fallacy.

Hoping those data science tips will help you in your carrier and became the best data scientist or data analyst in the company!

Source of material:

statistics mistakes,data science mistakes,statistical models,stats mistakes,bad models,machine learning models,overtraining models,data science tips,decision data,data errors,data science carrier,data analysis carrier,data analysis tips,data science advices,data scientist carrier,data analyst carrier,objective function,baseline model,