Black Lives Matter. Please consider donating to Black Girls Code today.

Slider enabling comparison of time series in variations (%)

Browsing the plotly functionalities advertised for finance I realise that there is a crucial feature missing .
(Note: I don’t want to sound critical - I find the recent improvements of plotly very impressive !)

What is missing ?
(At least I have found no example and not simple way to do it so far)
The ability to plot time series in relative value, with a slider over dates that plots the time series in percentage of their value on the left bound of the period selected by the slider.

See highcharts/highstock: http://www.highcharts.com/stock/demo/compare

I work in finance and I can guarantee that this is the most useful plot we have ever used.
To put is short, finance is all about time series (very often relative) values.

So much that I even wrapped highcharts for use inside a notebook:

Does the plotly team plan to add it to the fast growing capabilities of this interesting lib ?
Or is there already a way ? Some pointers ?

Thx

Hi there,
Have you seen our python tutorial on adding range sliders and range selectors to time series plots:

Yes, of course.
The configuration is expressive, and the result slick. No doubt !

But I haven’t found a way to plot the time series expressed in % of it value as of the left bound of the selected interval.

Let me rephrase.
Imagine I have 2 stocks (say AAPL as in your example, and also MSFT), and I want to see which one performed better over the interval selected by the slider.
How do I do that ? How do I compare them visually ?

This is what highcharts/highstocks enables you to do very easily (cf the link in my previous post).
I think a similar example should be posted - if it’s possible with plotly - is it ?

Thx

It’s been 4 years since this post, and I still can’t hardly find a solution. In finance we use terms like normalized or indexed to 100, unfortunately those terms have different meanings in python and pandas. It would seem that I can not be the only one looking for this solution.