Ok so I’ve been playing around with some parameters and trying to show the observed and predicted values of a timeseries data. The following are the values if you want to try it!

So far I have successfully created an area plot using df.plot(kind = “area”) method but I am stuck with setting the dates and legends of the figure.

I want to set “2022-07-27” as EQ and all values before EQ as negative values with 5 days interval for instance -30 -25 -20….. EQ 5 10 on xticks or as xticklabels.

For legend, I want to show “AT Obs.” vs “NARX Pred.”

this is the data I am currently working with.

https://drive.google.com/file/d/1RLQsFl-aKc6glHDZWY4xwpwJPbYwP0DH/view?usp=sharing

Here’s my current code and what I’ve been doing so far!

fig, ax = plt.subplots(figsize=(8,4))
at_day[["AT_Day_Time_Values", "NARX"]].plot(kind="area",stacked=False, ax=ax, color=["darkgreen", "red"])

Here’s one of my subplot which is basically sharing the x-axis with others