daveskender / stock.indicators Goto Github PK

Hi

Did you plan to add Hull Moving Average to your repository of indicators?

Thx
Ivan

Request to add Williams Fractal Indicator from @abcdamir2

Hi Dave,

One suggestion. Could you please add Weighted Moving average(WMA) and Hull Moving average(HMA). WMA is used to calculate HMA.

In IchomokuResults the correct name is Senkou, not Senkau

In v 0.10.10, Aroon does not pickup an edge case where High High was period + 1 days ago.
This is fixed in #125 and will be available in the next release.

hi
1-Stochastic Rsi usually return two value but your function return single value.
2- How to change %k and %d parameters in Stochastic Rsi?
Thanks

Hi

Is it possible to change inside Quote type of Volume from long to decimal? I know that for stocks long is fine, but when you are trading for example with crypto currencies volume is decimal number (as for example you can buy 0,00125 bitcoins).

Thank you
Ivan

Add Vortex Indicator, requested by @vebollu
Ref: https://en.wikipedia.org/wiki/Vortex_indicator

Please add Session Volume indicator [Volume Profile].

Here also one small suggestion: on couple of places you are using CultureInfo("en-US", false) - is it possible to make those properties public in order to adjust to local needs? I noticed it when I was cleaning history data and exception was thrown and had to figure out what was date behind exception text (as I'm from country where usual notation is dd.MM.yyyy). It's just cosmetic as it does not influence calculations.

From #110

I'm not going to criticize you work at all, it's cool that you put lot of effort in this project, I'm just offering some help if you want to talk about the design of the library and possible improvements both in memory and performance.

An example is about your Quote class, I think it's not a good idea to require a specific class because that way an user has to convert it's own models to Quote with problems both at performance level and design

A proper solution would be to use a IQuote interface, so that it's easy to adapt our own model because it's a matter of implementing that interface in our model without having to convert any object (most of the time our source model already contains close, high, low, date etc... so it's a no brainer to adapt)

Of course Quote class can still be there in the library for those that want to use that model.

Also the Index property is something not needed as I pointed out in a comment here #170 (comment)

but even if needed, having a different interface like IIndexedQuote required only by indicators that need to manage an index would be enough (and if we want to adapt our model adding the index property, we can do that with explicit interface declaration so that our model will not be polluted externally)

If you are open to changes I can try to find some time to contribute and do some benchmark too

Add .NET 5 framework targeting after it is released in mid-November 2020.

Hi,
i realy like your library, it works as described.

But i have a problem: when using the Stochastic Oscillator it upps my execution time by a lot. Going through 1000 stocks using other means of measurement takes 0.9 seconds (pretty fast machine and optimized code). Going through the same stocks using only the Stochastic Oscilator increases the time to 76 seconds.

This is how i call it:

 static public IEnumerable<StochResult> GetStochasticHistory(IEnumerable<Quote> history)
        {
            IEnumerable<StochResult> results = null;
            if (history.Count() > 8)
            {
                results = Indicator.GetStoch(history, 5, 3, 3);
            }
            return results;
        }

Measured with resharper, the "GetStoch" takes the time:

Is there any way to optimize the runtime? Is there maybe a light version?

Would it be possible to add the AROON indicators (upper and lower) as a function in this library?
I would very much appreciate it :)

I love your highly useful indicators project -- thank you for developing and sharing this. Amazing.

In any event, I'm running into this issue with both v1.2 and v1.3 NuGet packages:

I'm using a number of indicators such as GetStoch(), GetRsi(), etc. without issue. However, GetKama() is not present.

Brief example:

using Skender.Stock.Indicators;
...
... blah blah etc. etc.
...
... {
                var quoteHistory = tradeRepo.FindAllAsQuotes($"{stockName}_historic");
                
                var stoResults = Indicator.GetStoch(quoteHistory);   // works no issues
                var rsiResults = Indicator.GetRsi(quoteHistory); // works no issues
                var kemaResults = Indicator.GetKama(quoteHistory); // GetKama() not found
... }

Build result:

Error CS0117 'Indicator' does not contain a definition for 'GetKama'

Hi

I'm playing and testing this great library and so far I'm very pleased with it. I did noticed one thing dough - MACD calculation is rather slow. Here is an example:

• I'm using history of nearly 26000 quotes
  EMA: 30ms
  RSI: 70ms
  MACD: 11,3 sec
  BolingerBands: 10 sec

Are those expected times?

Ivan

Hi Dave,

Thanks, for this fantastic library you've created. Do you have any plans to add ALMA indicator to the library?

Thanks!

Before the first major release (see #104), I wanted to take some time to optimize the performance of the various indicators. I'm opening this issue to track and communicate the progress. This topic has come up in a couple of issues lately for people doing larger-scale operations (#88, #111).

As a baseline, here is the starting point on the performance of the current indicators using two years of historical daily stock quotes (e.g. 502 periods) and default or typical parameters.

I'm using BenchmarkDotNet as my performance benchmarking tool. I'll work on improving these numbers over the next few minor releases. Any help or feedback, as always, is appreciated.

For the latest results, see
https://dotnet.stockindicators.dev/performance

original baseline results

BenchmarkDotNet=v0.12.1, OS=Windows 10.0.19041.450 (2004/?/20H1)
Intel Core i7-7700HQ CPU 2.80GHz (Kaby Lake), 1 CPU, 8 logical and 4 physical cores
.NET Core SDK=3.1.401
  [Host]     : .NET Core 3.1.7 (CoreCLR 4.700.20.36602, CoreFX 4.700.20.37001), X64 RyuJIT
  DefaultJob : .NET Core 3.1.7 (CoreCLR 4.700.20.36602, CoreFX 4.700.20.37001), X64 RyuJIT

See 20-year-benchmarks-v0.10.6.xlsx for benchmarks using 20 years of daily history (5280 periods).

Before going to v1.0.0, I want to address a question that's been bothering me for a while. It has to do with the separation of concerns and how pure this library is regarding what it is and does. Early in development, I started adding additional fields to the output that I thought might be helpful, but am considering removing these additional fields. Some indicators (e.g. MACD, RSI, Heikin-Ashi) have basic analysis outputs, like IsBullish or IsIncreasing.

What kind of flourishes should we include in the library?

Take the Poll

For sure, I have no intention of turning this library into a trading system, so going so far as adding buy/sell signals is crossing the line (pun intended). Please also add additional comments below.

Yesterday I switched (for a brief moment) from TA-LIB to your approach, but in I'm getting divide errors. I'm using version version 1.1.2, is this anything familiar?

I think it seems to happen on coins with a low price (like viteusdt=0.0016), but I'm not yet sure that is actually the cause. I'll try to give you some test data soon (today i hope).

System.DivideByZeroException: Attempted to divide by zero.
at System.Decimal.FCallDivide(Decimal& d1, Decimal& d2)
at Skender.Stock.Indicators.Indicator.GetBollingerBands(IEnumerable`1 history, Int32 lookbackPeriod, Decimal standardDeviations)

List resultsbb = (List)Indicator.GetBollingerBands(history);

See AB#763. This also came up in #170 and #179, so it's probably a good time to explore removal of the internal Index.

Hi dave , thank you for the library you created and thank you for developing it . could you add 'support resistance' indicator in your library ?
thanks a lot .

Investigate the initialization values for the first reversal. Due to the "guess" on the initial trend, it may produce invalid initial results until the true trend emerges. This might be unavoidable; in that case, consider removing the first trend period (make values null).

Hi team.
I find SuperTrends result is not correct.
I tested it by "bitcoin 1 day candle 2017-2020" and compared with tradingview indicator result but not matched. (bollingerbands result is correct! good!)

can you check? thanks!

Private request from a user to add Tillson T3 Moving Average

Donchian Channel should evaluate the N periods prior to the current period, not including the current period, for high/low levels. It is currently using the current period, so no break-out would be identifiable.

add Choppiness Index

In a recent issue (#173) a user correctly identified that an indicator had less precision in an initialization phase for an indicator. Some indicators, like ADX, EMA, or others that use smoothing techniques require a runway of about 150-250 periods before the algorithm produces precision to 2 or more decimal places. Early initialization values can be imprecise to small 0.x to 0.0x decimals. This is not unusual for some indicators and precision to 0.00x is not always required for most use cases, especially when the user understands what is happening and can design around it by not using the first 150-250 values. Indicators that require a runway have this mentioned in the documentation.

Question: would you prefer that the library exclude values that don't meet a certain higher precision threshold or keep them? Excluding them would force users to use more history (and I'd probably update the Exception rules to enforce).

Thumbs up = enforce greater precision / exclude initialization values / require more history in Exception rules
Thumbs down = allow 0.x and 0.0x decimal imprecision in early periods and let me decide what to do

Feedback appreciated. Comment below with other ideas.

Hi,
First of all thanks for this awesome library.
Do you have any plan to create MFI indicator?

As many of you know, we're working towards a v1.0.0 version of this library. I've recently released v0.10.10 with a number of changes based on feedback from a poll (#92). While we will continue to add indicators before and after the v1.0.0 release, I'd like to consider this a major release candidate. At this point, I do not expect to add any significant breaking changes and view the currently published package (v0.11.5) as production-ready.

With that said, I have a few things I'd like to do before flipping the version number:

• Dogfood in my own trading system for a few weeks or more (in-progress)
• Add performance benchmarking - AB#727
• Further optimize performance of slower indicators (#114)
• Add indicator chart images to documentation - AB#717
• Incorporate any additional user concerns or ideas
• Maybe add a few more indicators

Feedback appreciated. Please comment below with your thoughts or submit issues for any problems. And if you want to help, let me know.

Hello
I'm trying your library (thanks for the effort so far) and I've found a bug regarding the usage of the quotes to generate different indicators using the same set of historical quotes but taking a different slice of them

Debug was hard on my side, so I took a look at your code and I see that your model classes have an internal Index that you use and I think that's the problem.

Suppose I get historical data from my POCO classes and convert them to your Quote model
suppose I've a list List<Quote> history of 100 Quotes

Now if I call

Indicator.GetBollingerBands(history.TakeLast(20), 20);
Indicator.GetRsi(history.TakeLast(14), 14);

I get a
System.ArgumentOutOfRangeException: 'Index was out of range. Must be non-negative and less than the size of the collection.'

I think the problem is that if I reuse Quote objects taking a different slice of the array, the bugs pops out because there is the Index field inconsistence inside your model classes

This means that historical quotes cannot be reused this way but I must always use the same set and I think it's not good because of performance (I'm not interested in applying indicators on the whole historical set in my scenario

Thanks

Hi guys
I want to use EMA with lookbackPeriod=101 but I got an error that says
"
$exception | {"Insufficient history provided for EMA. You provided 101 periods of history when at least 202 is required. Since this uses a smoothing technique, for a lookback period of 101, we recommend you use at least 351 data points prior to the intended usage date for maximum precision."} | Skender.Stock.Indicators.BadHistoryException
"
P.S: If I set any lookbackPeriod but I got a similar error

In v0.10.10 the CMF output is not on the latest scaling model.
This was fixed in #126 and will be available in the next release.

Implement Williams Alligator
https://www.investopedia.com/articles/trading/072115/exploring-williams-alligator-indicator.asp

[private] user request

Hi Dave,

First, thank you for your contribution to this project. I am excited to find this time-saving project a couple of weeks ago.
The issue I am submitting has to do with Average Directional Index (ADX). I notice that it is somewhat off when I compare it with tradingview's ADX/DMI indicator.

For example, on 2000-11-27, the Skender ADX indicator shows
ADX=30.1245
+DI=16.898
-DI=30.579

On the same day, TradingView indicator shows
ADX=30.4113
+DI=16.9263
-DI=30.6871

Could you check if this is a bug?

Thank you,

I'm getting my historical data from yahoo finance, I have ensured everything is present in the Quote class, and then I pass the data into

int fastPeriod = 12; int slowPeriod = 26; int signalPeriod = 9; IEnumerable<MacdResult> results = Indicator.GetMacd(history, fastPeriod, slowPeriod, signalPeriod);
But it gives me an exception on this line, every time saying "No historical quote provided".

I also have a stack trace for you if that's helpful:

at Skender.Stock.Indicators.Cleaners.PrepareHistory(IEnumerable1 history)\r\n at Skender.Stock.Indicators.Indicator.GetEma(IEnumerable1 history, Int32 lookbackPeriod)\r\n at Skender.Stock.Indicators.Indicator.GetMacd(IEnumerable1 history, Int32 fastPeriod, Int32 slowPeriod, Int32 signalPeriod)\r\n at PT.Middleware.Indicators.GetIndicatorComposite(String symbol, String function, IEnumerable1 history, Int32 daysToCalculate) in C:\\Users\\linki\\Dev\\Pro-Trades\\Middleware\\Indicators.cs:line 55

I tried preparing it with the Cleaners class but it is still giving me the same exception. Please tell me what the fix is or otherwise fix this bug?

Thanks a bunch!

I'm using your nuget package from a month , it's been really good. I have the following issues with the package.

1. OnBalanceVolume Indicator.GetObv(history) is not match with stockcharts/tradingview obv value
2. Accumulation Distribution (ADL) Indicator.GetAdl(history) also not matching with the stockcharts/tradingview Adl value

Suggestions:

Earlier we have properties IsReversal, Isrising, Isdiverging in MAcd, Rsi, PsarIsrising etc... can you please add those back if possible? can you please apply the same for aaron and others?

I'm using this API for [quote] data (https://docs.coinapi.io/#historical-data) but I can't understand why what I get from these functions (TEMA) from Skender.Stock.Indicators are so different from what I see in the Binance.

Hi Dave!

Good work on the library. I am finding it very useful in my own project and I would be glad to contribute to this library once I am finished with my project. Until then, I would like to suggest adding these results in the output of SMA.

• Mean Absolute Deviation
• Mean Square Error
• Mean Absolute Percentage Error

As referenced in this discussion:

Community discussion about Slope and Linear Regression

Originally posted by @DaveSkender in #241

Recommend by @Fredrik-C while discussing MAMA, which is a similar formula.

Could you please add the TD sequential indicator?

https://www.investopedia.com/terms/d/demarkerindicator.asp

Thanks and regards,
Marius

GetKlinesBefore_300(DateTime.Now);
private void GetKlinesBefore_300(DateTime dateTime)
{
init_quote_300_Tick_For_Signals.AddRange(BiClient.Spot.Market.GetKlines(pair, KlineInterval.ThreeMinutes, endTime: dateTime)
.Data.Select(u => new Quote
{
Open = u.Open,
Close = u.Close,
Low = u.Low,
High = u.High,
Volume = u.QuoteVolume,
Date = u.CloseTime
}).ToList());
}
decimal? RSI = Indicator.GetRsi(init_quote_300_Tick_For_Signals, 4).LastOrDefault().Rsi;
decimal? RSI_24 = Indicator.GetRsi(init_quote_300_Tick_For_Signals, 6).LastOrDefault().Rsi;
var AROON = Indicator.GetAroon(init_quote_300_Tick_For_Signals,23).LastOrDefault();

using System;
using System.Collections.Generic;
using System.Linq;
using Skender.Stock.Indicators;

namespace Stock_API_Core.Calculations
{
public class Rsi
{
public void sample()
{
// fetch historical quotes from your favorite feed, in Quote format
IEnumerable history = GetHistoryFromFeed("MSFT");

        // calculate 20-period SMA
        IEnumerable<SmaResult> results = Indicator.GetSma(history,20);

        // use results as needed
        SmaResult result = results.LastOrDefault();
        Console.WriteLine("SMA on {0} was ${1}", result.Date, result.Sma);
    }
}

}

The GetHistoryFromFeed("MSFT"); isn't being recognized despite importing Skender.Stock.Indicators

I don't understand why this is happening

Based on feedback from the poll (#92), remove the Index value from Quote and various xxxResult classes. Since these are used by the library, we need to keep them; however, make them internal only for simplification and to make it easier for users to extend classes without having to deal with protection level on that property. We'll need to verify that pre-cleaning still works as intended to improve performance for users who are doing bulk operations on the same history.

Based on feedback from the poll (#92), remove the various superfluous flourished from indicator output classes, such as IsIncreasing and IsBullish. Only deliver core indicator values.

Add Fractal Chaos Bands, requested by "BA".

Related to #183, there are a few other indicators that are vulnerable to divide by zero errors when unusual historical quotes are provided; for example, when price does not change or there are zero values. AB#775

Hello Dave Skender
One more suggestion can you add a Custom RSI indicator which is a very demanding study indicator.
Reference : https://www.stockfetcher.com/forums/Indicators/Custom-RSI-CRSI/112583
To CRSI we can pass other indicator output too