Know your data, the fundamentals of analysis
Thanks to Ben Hunt's newest monetary letter, I pulled out my vintage copies of work by Ed Tufte, the visible information extraordinaire. He might have performed extra for nice visualization and information research than anybody over the final 40 years. The Anscombe Quartet says all of it in the event you do not seem to be on the data.
I took the Tufte at some point course on the visualization of information over 20 years ago and it has had a profound impact on focusing my powers of observation and my capacity to suppose of blank representations of data. Over the years, I actually have adopted a set of guidelines while watching at information that has served me well.
My information rule checklist:
1. Graph the information - If you can't see it, you can't recognize it. Pictures are critical.
2. Cut the information - Look over unique time intervals to see if there are changes. Data move.
3. Get the stats - Understand the distribution of the information in simple terms after seeing it first.
4. Look on the histogram - Find for outliers simply due to the fact it really is had been funds is won or lost.
5. Match information with occasions - Go equally methods - match outliers with occasions and large occasions with data.
6. Generate conditional tables - The energy of a 2x2 contingency desk is typically missed.
7. Check the volatility and correlation - See regardless of even if volatility and correlations change.
8. Compare towards nice occasions and bad - Show industry cycle and large occasions on graph.
9. Map idea and past proof - Use priors of unique researchers to focus representation of data.
10. Run your regression assessments - Be skeptical and discuss residuals.
Good information research is a talent that might be realized by way of following a process.
I took the Tufte at some point course on the visualization of information over 20 years ago and it has had a profound impact on focusing my powers of observation and my capacity to suppose of blank representations of data. Over the years, I actually have adopted a set of guidelines while watching at information that has served me well.
My information rule checklist:
1. Graph the information - If you can't see it, you can't recognize it. Pictures are critical.
2. Cut the information - Look over unique time intervals to see if there are changes. Data move.
3. Get the stats - Understand the distribution of the information in simple terms after seeing it first.
4. Look on the histogram - Find for outliers simply due to the fact it really is had been funds is won or lost.
5. Match information with occasions - Go equally methods - match outliers with occasions and large occasions with data.
6. Generate conditional tables - The energy of a 2x2 contingency desk is typically missed.
7. Check the volatility and correlation - See regardless of even if volatility and correlations change.
8. Compare towards nice occasions and bad - Show industry cycle and large occasions on graph.
9. Map idea and past proof - Use priors of unique researchers to focus representation of data.
10. Run your regression assessments - Be skeptical and discuss residuals.
Good information research is a talent that might be realized by way of following a process.