adelorenzo/stable
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Add Backtesting and Hyperopt documentation

Browse Source
This commit is contained in:
Gerald Lonlas 2018-01-02 22:43:00 -08:00
parent f37c495b90
commit cb7c36a512
4 changed files with 379 additions and 47 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

104
docs/backtesting.md Normal file
View File

@ -0,0 +1,104 @@
# Backtesting
This page explains how to validate your strategy performance by using
Backtesting.
## Table of Contents
- [Test your strategy with Backtesting](#test-your-strategy-with-backtesting)
- [Understand the backtesting result](#understand-the-backtesting-result)
## Test your strategy with Backtesting
Now you have good Buy and Sell strategies, you want to test it against
real data. This is what we call
[backtesting](https://en.wikipedia.org/wiki/Backtesting).
Backtesting will use the crypto-currencies (pair) from your config file
and load static tickers located in
[/freqtrade/tests/testdata](https://github.com/gcarq/freqtrade/tree/develop/freqtrade/tests/testdata).
If the 5 min and 1 min ticker for the crypto-currencies to test is not
already in the `testdata` folder, backtesting will download them
automatically. Testdata files will not be updated until you specify it.
The result of backtesting will confirm you if your bot as more chance to
make a profit than a loss.
The backtesting is very easy with freqtrade.
### Run a backtesting against the currencies listed in your config file
**With 5 min tickers (Per default)**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation
```
**With 1 min tickers**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --ticker-interval 1
```
**Reload your testdata files**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --refresh-pairs-cached
```
**With live data (do not alter your testdata files)**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --live
```
For help about backtesting usage, please refer to
[Backtesting commands](#backtesting-commands).
## Understand the backtesting result
The most important in the backtesting is to understand the result.
A backtesting result will look like that:
```
====================== BACKTESTING REPORT ================================
pair buy count avg profit % total profit BTC avg duration
-------- ----------- -------------- ------------------ --------------
BTC_ETH 56 -0.67 -0.00075455 62.3
BTC_LTC 38 -0.48 -0.00036315 57.9
BTC_ETC 42 -1.15 -0.00096469 67.0
BTC_DASH 72 -0.62 -0.00089368 39.9
BTC_ZEC 45 -0.46 -0.00041387 63.2
BTC_XLM 24 -0.88 -0.00041846 47.7
BTC_NXT 24 0.68 0.00031833 40.2
BTC_POWR 35 0.98 0.00064887 45.3
BTC_ADA 43 -0.39 -0.00032292 55.0
BTC_XMR 40 -0.40 -0.00032181 47.4
TOTAL 419 -0.41 -0.00348593 52.9
```
The last line will give you the overall performance of your strategy,
here:
```
TOTAL 419 -0.41 -0.00348593 52.9
```
We understand the bot has made `419` trades for an average duration of
`52.9` min, with a performance of `-0.41%` (loss), that means it has
lost a total of `-0.00348593 BTC`.
As you will see your strategy performance will be influenced by your buy
strategy, your sell strategy, and also by the `minimal_roi` and
`stop_loss` you have set.
As for an example if your minimal_roi is only `"0": 0.01`. You cannot
expect the bot to make more profit than 1% (because it will sell every
time a trade will reach 1%).
```json
"minimal_roi": {
"0": 0.01
},
```
On the other hand, if you set a too high `minimal_roi` like `"0": 0.55`
(55%), there is a lot of chance that the bot will never reach this
profit. Hence, keep in mind that your performance is a mix of your
strategies, your configuration, and the crypto-currency you have set up.
## Next step
Great, your strategy is profitable. What if the bot can give your the
optimal parameters to use for your strategy?
Your next step is to learn [how to find optimal parameters with Hyperopt](https://github.com/gcarq/freqtrade/blob/develop/docs/hyperopt.md)

48
docs/bot-optimization.md
View File

@ -1,14 +1,10 @@
# Bot Optimization # Bot Optimization
This page explains where to customize your strategies, validate their This page explains where to customize your strategies, and add new
performance by using Backtesting, and tuning them by finding the optimal indicators.
parameters with Hyperopt.
## Table of Contents ## Table of Contents
- [Change your strategy](#change-your-strategy) - [Change your strategy](#change-your-strategy)
- [Add more Indicator](#add-more-indicator) - [Add more Indicator](#add-more-indicator)
- [Test your strategy with Backtesting](#test-your-strategy-with-backtesting)
- [Find optimal parameters with Hyperopt](#find-optimal-parameters-with-hyperopt)
- [Show your buy strategy on a graph](#show-your-buy-strategy-on-a-graph)
## Change your strategy ## Change your strategy
The bot is using buy and sell strategies to buy and sell your trades. The bot is using buy and sell strategies to buy and sell your trades.
@ -115,43 +111,7 @@ def populate_indicators(dataframe: DataFrame) -> DataFrame:
return dataframe return dataframe
``` ```
## Test your strategy with Backtesting
Now you have good Buy and Sell strategies, you want to test it against
real data. This is what we call [backtesting](https://en.wikipedia.org/wiki/Backtesting).
Backtesting will use the crypto-currencies (pair) tickers located in
[/freqtrade/tests/testdata](https://github.com/gcarq/freqtrade/tree/develop/freqtrade/tests/testdata).
If the 5 min and 1 min ticker for the crypto-currencies to test is not
already in the `testdata` folder, backtesting will download them
automatically. Testdata files will not be updated until your specify it.
### Run a backtesting against the currencies listed in your config file
**With 5 min tickers (Per default)**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation
```
**With 1 min tickers**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --ticker-interval 1
```
**Reload your testdata files**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --refresh-pairs-cached
```
**With live data (do not alter your testdata files)**
```bash
python3 ./freqtrade/main.py backtesting --realistic-simulation --live
```
## Find optimal parameters with Hyperopt
*To be completed, please feel free to complete this section.*
## Show your buy strategy on a graph
*To be completed, please feel free to complete this section.*
## Next step ## Next step
Now you have a perfect bot and want to control it from Telegram. Your Now you have a perfect strategy you probably want to backtesting it.
next step is to learn [Telegram usage](https://github.com/gcarq/freqtrade/blob/develop/docs/telegram-usage.md). Your next step is to learn [How to use ](https://github.com/gcarq/freqtrade/blob/develop/docs/backtesting.md).

269
docs/hyperopt.md Normal file
View File

@ -0,0 +1,269 @@
# Hyperopt
This page explains how to tune your strategy by finding the optimal
parameters with Hyperopt.
## Table of Contents
- [Prepare your Hyperopt](#prepare-hyperopt)
- [1. Configure your Guards and Triggers](#1-configure-your-guards-and-triggers)
- [2. Update the hyperopt config file](#2-update-the-hyperopt-config-file)
- [Advanced Hyperopt notions](#advanced-notions)
- [Understand the Guards and Triggers](#understand-the-guards-and-triggers)
- [Execute Hyperopt](#execute-hyperopt)
- [Hyperopt with MongoDB](#hyperopt-with-mongoDB)
- [Understand the hyperopts result](#understand-the-backtesting-result)
## Prepare Hyperopt
Before we start digging in Hyperopt, we recommend you to take a look at
out Hyperopt file
[freqtrade/optimize/hyperopt.py](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py)
### 1. Configure your Guards and Triggers
There are two places you need to change to add a new buy strategy for
testing:
- Inside the [populate_buy_trend()](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py#L167-L207).
- Inside the [SPACE dict](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py#L47-L94).
There you have two different type of indicators: 1. `guards` and 2.
`triggers`.
1. Guards are conditions like "never buy if ADX < 10", or never buy if
current price is over EMA10.
2. Triggers are ones that actually trigger buy in specific moment, like
"buy when EMA5 crosses over EMA10" or buy when close price touches lower
bollinger band.
HyperOpt will, for each eval round, pick just ONE trigger, and possibly
multiple guards. So that the constructed strategy will be something like
"*buy exactly when close price touches lower bollinger band, BUT only if
ADX > 10*".
If you have updated the buy strategy, means change the content of
`populate_buy_trend()` function you have to update the `guards` and
`triggers` hyperopts must used.
As for an example if your `populate_buy_trend()` function is:
```python
def populate_buy_trend(dataframe: DataFrame) -> DataFrame:
dataframe.loc[
(dataframe['rsi'] < 35) &
(dataframe['adx'] > 65),
'buy'] = 1
return dataframe
```
Your hyperopt file must contains `guards` to find the right value for
`(dataframe['adx'] > 65)` & and `(dataframe['plus_di'] > 0.5)`. That
means you will need to enable/disable triggers.
In our case the `SPACE` and `populate_buy_trend` in hyperopt.py file
will be look like:
```python
SPACE = {
'rsi': hp.choice('rsi', [
{'enabled': False},
{'enabled': True, 'value': hp.quniform('rsi-value', 20, 40, 1)}
]),
'adx': hp.choice('adx', [
{'enabled': False},
{'enabled': True, 'value': hp.quniform('adx-value', 15, 50, 1)}
]),
'trigger': hp.choice('trigger', [
{'type': 'lower_bb'},
{'type': 'faststoch10'},
{'type': 'ao_cross_zero'},
{'type': 'ema5_cross_ema10'},
{'type': 'macd_cross_signal'},
{'type': 'sar_reversal'},
{'type': 'stochf_cross'},
{'type': 'ht_sine'},
]),
}
...
def populate_buy_trend(dataframe: DataFrame) -> DataFrame:
conditions = []
# GUARDS AND TRENDS
if params['adx']['enabled']:
conditions.append(dataframe['adx'] > params['adx']['value'])
if params['rsi']['enabled']:
conditions.append(dataframe['rsi'] < params['rsi']['value'])
# TRIGGERS
triggers = {
'lower_bb': dataframe['tema'] <= dataframe['blower'],
'faststoch10': (crossed_above(dataframe['fastd'], 10.0)),
'ao_cross_zero': (crossed_above(dataframe['ao'], 0.0)),
'ema5_cross_ema10': (crossed_above(dataframe['ema5'], dataframe['ema10'])),
'macd_cross_signal': (crossed_above(dataframe['macd'], dataframe['macdsignal'])),
'sar_reversal': (crossed_above(dataframe['close'], dataframe['sar'])),
'stochf_cross': (crossed_above(dataframe['fastk'], dataframe['fastd'])),
'ht_sine': (crossed_above(dataframe['htleadsine'], dataframe['htsine'])),
}
...
```
### 2. Update the hyperopt config file
Hyperopt is using a dedicated config file. At this moment hyperopt
cannot use your config file. It is also made on purpose to allow you
testing your strategy with different configurations.
The Hyperopt configuration is located in
[freqtrade/optimize/hyperopt_conf.py](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt_conf.py).
## Advanced notions
### Understand the Guards and Triggers
When you need to add the new guards and triggers to be hyperopt
parameters, you do this by adding them into the [SPACE dict](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py#L47-L94).
If it's a trigger, you add one line to the 'trigger' choice group and that's it.
If it's a guard, you will add a line like this:
```
'rsi': hp.choice('rsi', [
{'enabled': False},
{'enabled': True, 'value': hp.quniform('rsi-value', 20, 40, 1)}
]),
```
This says, "*one of guards is RSI, it can have two values, enabled or
disabled. If it is enabled, try different values for it between 20 and 40*".
So, the part of the strategy builder using the above setting looks like
this:
```
if params['rsi']['enabled']:
conditions.append(dataframe['rsi'] < params['rsi']['value'])
```
It checks if Hyperopt wants the RSI guard to be enabled for this
round `params['rsi']['enabled']` and if it is, then it will add a
condition that says RSI must be < than the value hyperopt picked
for this evaluation, that is given in the `params['rsi']['value']`.
That's it. Now you can add new parts of strategies to Hyperopt and it
will try all the combinations with all different values in the search
for best working algo.
### Add a new Indicators
If you want to test an indicator that isn't used by the bot currently,
you need to add it to
[freqtrade/analyze.py](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/analyze.py#L40-L70)
inside the `populate_indicators` function.
## Execute Hyperopt
Once you have updated your hyperopt configuration you can run it.
Because hyperopt tries a lot of combination to find the best parameters
it will take time you will have the result (more than 30 mins).
We strongly recommend to use `screen` to prevent any connection loss.
```bash
python3 ./freqtrade/main.py -c config.json hyperopt
```
### Hyperopt with MongoDB
Hyperopt with MongoDB, is like Hyperopt under steroids. As you saw by
executing the previous command is the execution takes a long time.
To accelerate it you can use hyperopt with MongoDB.
To run hyperopt with MongoDb you will need 3 terminals.
**Terminal 1: Start MongoDB**
```bash
cd <freqtrade>
source .env/bin/activate
python3 scripts/start-mongodb.py
```
**Terminal 2: Start Hyperopt worker**
```bash
cd <freqtrade>
source .env/bin/activate
python3 scripts/start-hyperopt-worker.py
```
**Terminal 3: Start Hyperopt with MongoDB**
```bash
cd <freqtrade>
source .env/bin/activate
python3 ./freqtrade/main.py -c config.json hyperopt --use-mongodb
```
**Re-run an Hyperopt**
To re-run Hyperopt you have to delete the existing MongoDB table.
```bash
cd <freqtrade>
rm -rf .hyperopt/mongodb/
```
## Understand the hyperopts result
Once Hyperopt is completed you can use the result to adding new buy
signal. Given following result from hyperopt:
```
Best parameters:
{
"adx": 1,
"adx-value": 15.0,
"fastd": 1,
"fastd-value": 40.0,
"green_candle": 1,
"mfi": 0,
"over_sar": 0,
"rsi": 1,
"rsi-value": 37.0,
"trigger": 0,
"uptrend_long_ema": 1,
"uptrend_short_ema": 0,
"uptrend_sma": 0
}
Best Result:
2197 trades. Avg profit 1.84%. Total profit 0.79367541 BTC. Avg duration 241.0 mins.
```
You should understand this result like:
- You should **consider** the guard "adx" (`"adx": 1,` = `adx` is true)
and the best value is `15.0` (`"adx-value": 15.0,`)
- You should **consider** the guard "fastd" (`"fastd": 1,` = `fastd`
is true) and the best value is `40.0` (`"fastd-value": 40.0,`)
- You should **consider** to enable the guard "green_candle"
(`"green_candle": 1,` = `candle` is true) but this guards as no
customizable value.
- You should **ignore** the guard "mfi" (`"mfi": 0,` = `mfi` is false)
- and so on...
You have to look from
[freqtrade/optimize/hyperopt.py](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py#L170-L200)
what those values match to.
So for example you had `adx-value: 15.0` (and `adx: 1` was true) so we
would look at `adx`-block from
[freqtrade/optimize/hyperopt.py](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/optimize/hyperopt.py#L178-L179).
That translates to the following code block to
[analyze.populate_buy_trend()](https://github.com/gcarq/freqtrade/blob/develop/freqtrade/analyze.py#L73)
```
(dataframe['adx'] > 15.0)
```
So translating your whole hyperopt result to as the new buy-signal
would be the following:
```
def populate_buy_trend(dataframe: DataFrame) -> DataFrame:
dataframe.loc[
(
(dataframe['adx'] > 15.0) & # adx-value
(dataframe['fastd'] < 40.0) & # fastd-value
(dataframe['close'] > dataframe['open']) & # green_candle
(dataframe['rsi'] < 37.0) & # rsi-value
(dataframe['ema50'] > dataframe['ema100']) # uptrend_long_ema
),
'buy'] = 1
return dataframe
```
## Next step
Now you have a perfect bot and want to control it from Telegram. Your
next step is to learn the [Telegram usage](https://github.com/gcarq/freqtrade/blob/develop/docs/telegram-usage.md).

5
docs/index.md
View File

@ -22,9 +22,8 @@ Pull-request. Do not hesitate to reach us on
- [Bot Optimization](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md) - [Bot Optimization](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md)
- [Change your strategy](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#change-your-strategy) - [Change your strategy](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#change-your-strategy)
- [Add more Indicator](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#add-more-indicator) - [Add more Indicator](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#add-more-indicator)
- [Test your strategy with Backtesting](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#test-your-strategy-with-backtesting) - [Test your strategy with Backtesting](https://github.com/gcarq/freqtrade/blob/develop/docs/backtesting.md)
- [Find optimal parameters with Hyperopt](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#find-optimal-parameters-with-hyperopt) - [Find optimal parameters with Hyperopt](https://github.com/gcarq/freqtrade/blob/develop/docs/hyperopt.md)
- [Show your buy strategy on a graph](https://github.com/gcarq/freqtrade/blob/develop/docs/bot-optimization.md#show-your-buy-strategy-on-a-graph)
- [Control the bot with telegram](https://github.com/gcarq/freqtrade/blob/develop/docs/telegram-usage.md) - [Control the bot with telegram](https://github.com/gcarq/freqtrade/blob/develop/docs/telegram-usage.md)
- [Contribute to the project](https://github.com/gcarq/freqtrade/blob/develop/CONTRIBUTING.md) - [Contribute to the project](https://github.com/gcarq/freqtrade/blob/develop/CONTRIBUTING.md)
- [How to contribute](https://github.com/gcarq/freqtrade/blob/develop/CONTRIBUTING.md) - [How to contribute](https://github.com/gcarq/freqtrade/blob/develop/CONTRIBUTING.md)