Elder's market thermometer calculation

First of all, thanks for the work on this library! I have a stupid question about the Elder’s Market Thermometer indicator. I’ve been using the indicator from both Tradingview (using lazybear’s version) and from Python using pandas_ta. I noticed the values are different between the two versions, is it a bug or is it a difference in the calculation?

Here is an example with BTCUSDT 4h (Binance):

Tradingview:

And here is what i tried from my code:

thermo = df.ta.thermo(length=22, long=3, short=7)
>>print(thermo

              THERMO_22_3.0_7.0  THERMOma_22_3.0_7.0
...
2021-04-18 00:00:00            9038.68          1415.982374
2021-04-18 04:00:00            3971.96          1638.241298
2021-04-18 08:00:00             860.10          1570.576837
2021-04-18 12:00:00             723.77          1496.941460
2021-04-18 16:00:00            1659.75          1511.098724
2021-04-18 20:00:00             720.62          1442.361444
2021-04-19 00:00:00             602.72          1369.349144
2021-04-19 04:00:00             720.06          1312.889219
2021-04-19 08:00:00             290.65          1223.998852
2021-04-19 12:00:00            1898.45          1282.646778
2021-04-19 16:00:00            1042.20          1261.738362
2021-04-19 20:00:00            1165.78          1253.394157
2021-04-20 00:00:00            1280.41          1255.743361
2021-04-20 04:00:00             907.08          1225.424808
2021-04-20 08:00:00             932.82          1199.980911
2021-04-20 12:00:00             419.20          1132.086919
2021-04-20 16:00:00             450.05          1072.779361
2021-04-20 20:00:00             990.37          1065.613329
2021-04-21 00:00:00             919.39          1052.898257
2021-04-21 04:00:00             935.14          1042.658409
2021-04-21 08:00:00             691.84          1012.152460
2021-04-21 12:00:00             596.70           976.026159
2021-04-21 16:00:00             829.00           963.241276
2021-04-21 20:00:00            1592.98          1018.001165
2021-04-22 00:00:00             975.02          1014.263672
2021-04-22 04:00:00             889.00          1003.371179
2021-04-22 08:00:00             373.00           948.556294
2021-04-22 12:00:00            1311.56           980.121834

I’m getting the OHLC data from here: https://api.binance.com/api/v1/klines?&symbol=BTCUSDT&interval=4h&limit=1200

Here, for example, the last candle will have a value of 1311 from Python, while according to Lazybear’s version is around 300, the same happens with other candles. Now, am i missing something and this is indeed a stupid question or is the calculation just different? Thanks in advance!

[Jacks349]

Hello @Jacks349,

Hmm… that’s weird. Thanks for providing details and source.

Also if you have not come across it already, here is some discussion about the creation of thermo in Issue #119.

When I test TV indicators, I export from them to do correlation testing. If you could do that when you get a chance and print a correlation result, that would be great.

If you have a TradingView account, add LazyBear’s Thermo indicator to a new basic chart. Just use ticker SPY and Daily timeframe. So the new chart should mostly be: the Candles, Volume and Volume MA, and LazyBear’s Thermo Plots. Zoom out as far as you can and then you should read Issue #107 on how to export data from TradingView and do correlation testing. That will give you a good sample to compare visually. Anywhere between 500 and 1000 rows is good. Plot the data on the last 100 or so bars to see how it looks and 300-400 bars for a bigger picture to make sure it looks consistent on a large perspective. You can test Crypto later once you are certain it looks about right or its nearly 99% correlated.

At this time, I currently do not have an expected time on when I can take on this Issue. But I’ll keep you posted.

Kind Regards, KJ

I’m going to try to do correlation testing, once i know how to do it properly, thank you! I’m trying to understand how does Lazybear calculates it

[twopirllc]

@Jacks349,

Sorry for the late answer! I did not manage to do the test due to a lack of time, but i plan to do it. You can close the issue for now, at worst i’ll reopen it and post my results!

No worries. Thanks for getting back to me. I’ll close for now and you can reopen when ready.

KJ