2019-01-23 | SNIA PM Summit Hackathon January | Runtime Evironment library

I have a pivot table that is intermediate between users and subscriptions. Here is the schema:

I try to Generate Monthly KPIs DataFrame by aggregating Revenue but i got this error :

AttributeError: 'DatetimeIndex' object has no attribute 'Year_Lease_Start'

Here is the format of the dataframe :

I made this code where the purpose its to retrieve data from an API from a certain number of stock tickers

Overview

I am trying to use this Python Zig Zag candlestick indicator (utilises High,Low,Close values) on financial data but the code below appears to have a bug.

Any help fixing this is appreciated or if there is another working Python module that provides this functionality please advise.

What is a Zig Zag indicator

"The Zig Zag indicator plots points on the chart whenever prices reverse by a percentage greater than a pre-chosen variable." Source

What have I tried

While searching for a Python zigzag indicator for candlestick charts the only code I could find was from this pull request.

I have a timeseries df with 5 years of stock index prices (so df has 2 columns Date and Price. I then have a new column '3M forward date' which is just the original Date column + 3 months.

I'm then trying to create a '3M forward Price' column. This would be the corresponding price from the original Price column but at the 3M forward date.

I need to find the syntax for saying : Find the date from the 3M Forward Date column within the Date column and append the corresponding value from the original Price column toa new 3M forward price column.

I've tried variation of .loc ideas and looked through several historic questions but given my lookup value is from a df column and requiring the appending to a new column I cant figure it out.

Small section of current DF :

I have the following table with date range:

I've asked this question previously (Identfy breaks in time series and assign unique factor for each break in R) and the solution suggested worked at the time, but I now see it does not work in all cases. I've been trialing variations of that solution but have had no luck. Here is the simplified question again:

I have a date-time series of vessel positions, with large gaps in the time series. Gaps represent breaks in the vessel track. I want to add a unique identifier to each track. Here is some real data;

I have uploaded a xlsx file into python with this as an output: