Correlation coefficient is a measure of the strength of the relationship between two values. It can be useful for market analysis, cryptocurrencies, forex and much more.
Since it "describes the degree to which two series tend to deviate from their moving average values" (1), first of all you have to set the length of these moving averages. You can also retrieve the values from another timeframe, and choose whether or not to ignore the gaps.
After selecting the reference ticker, which is not dependent from the chart you are on, you can choose up to eight other tickers to relate to it. The provided matrix table will then give you a deeper insight through all of the correlations between the chosen symbols.
Correlation values are scored on a scale from 1 to -1
A value of 1 means the correlation between the values is perfect.
A value of 0 means that there is no correlation at all.
A value of -1 indicates that the correlation is perfectly opposite.
For a better view at a glance, eight level colors are available and it is possible to modify them at will. You can even change level ranges by setting their threshold values. The background color of the matrix's cells will change accordingly to all of these choices.
The default threshold values, commonly used in statistics, are as follows:
None to weak correlation: 0 - 0.3
Weak to moderate correlation: 0.3 - 0.5
Moderate to high correlation: 0.5 - 0.7
High to perfect correlation: 0.7 - 1
Remember to be careful about spurious correlations, which are strong correlations without a real causal relationship.
(1) www.tradingview.com/...script-reference/v5/
Since it "describes the degree to which two series tend to deviate from their moving average values" (1), first of all you have to set the length of these moving averages. You can also retrieve the values from another timeframe, and choose whether or not to ignore the gaps.
After selecting the reference ticker, which is not dependent from the chart you are on, you can choose up to eight other tickers to relate to it. The provided matrix table will then give you a deeper insight through all of the correlations between the chosen symbols.
Correlation values are scored on a scale from 1 to -1
A value of 1 means the correlation between the values is perfect.
A value of 0 means that there is no correlation at all.
A value of -1 indicates that the correlation is perfectly opposite.
For a better view at a glance, eight level colors are available and it is possible to modify them at will. You can even change level ranges by setting their threshold values. The background color of the matrix's cells will change accordingly to all of these choices.
The default threshold values, commonly used in statistics, are as follows:
None to weak correlation: 0 - 0.3
Weak to moderate correlation: 0.3 - 0.5
Moderate to high correlation: 0.5 - 0.7
High to perfect correlation: 0.7 - 1
Remember to be careful about spurious correlations, which are strong correlations without a real causal relationship.
(1) www.tradingview.com/...script-reference/v5/
版本注释:
Minor fix: The transparency setting affects borders too.
版本注释:
Added an option to show the correlation length on the corner cell.
Cleaned the code.
Cleaned the code.
版本注释:
Minor update:
1) Corrected the letter of the correlation value (which now is ' r '), since it refers to a sample and not a population.
2) Cleaned the code and improved its overall appearance.
1) Corrected the letter of the correlation value (which now is ' r '), since it refers to a sample and not a population.
2) Cleaned the code and improved its overall appearance.