help to use the analyzer

[ZzzzHeJ]

I want to use analysis tools to draw daily revenue curves and statistics Sharp. I set timer and add_return every day, but the returns will reset when the backtest is over. What's the right thing to do please.

            venue_positions = []
            venue_currencies = set()
            for position in positions:
                if position.instrument_id.venue == venue.id:
                    venue_positions.append(position)
                    venue_currencies.add(position.quote_currency)
                    if position.base_currency is not None:
                        venue_currencies.add(position.base_currency)

            # Calculate statistics
            self._kernel.portfolio.analyzer.calculate_statistics(account, venue_positions)

[cjdsellers]

Hi @ZzzzHeJ

Thanks for reaching out.

All data remains available until engine.reset() is called on the engine.

[ZzzzHeJ]

I just run

node = BacktestNode(configs=[config])

result = node.run()

it seems like will auto call engine.reset() in the run().
please tell me what should I do, and the best practice to get the daliy return

[cjdsellers]

Hi @ZzzzHeJ

You could try passing a config with dispose_on_completion of False, and see if you can access the data from the analyzer as you expect.

[ZzzzHeJ]

Thanks, I will try it, by the way, I found a bug in the analyzer

nautilus_trader/nautilus_trader/analysis/analyzer.py

Line 146 in b65c810

self._returns.sort_index()

the sort_index() is not effect the self._return, maybe use self._return = self._return .sort_index()?

[ZzzzHeJ]

Unfortunately, it doesn't work, because the analyzer._returns will be reset in the calculate_statistics

nautilus_trader/nautilus_trader/analysis/analyzer.py

Lines 128 to 146 in b65c810

	def calculate_statistics(self, account: Account, positions: list[Position]) -> None:
	"""
	Calculate performance metrics from the given data.
	Parameters
	----------
	account : Account
	The account for the calculations.
	positions : list[Position]
	The positions for the calculations.
	"""
	self._account_balances_starting = account.starting_balances()
	self._account_balances = account.balances_total()
	self._realized_pnls = {}
	self._returns = pd.Series(dtype=float64)
	self.add_positions(positions)
	self._returns.sort_index()

This is the call stack analyzer.calculate_statistics<-engine._log_post_run<-engine.end, the end() will call after _run() with only streaming judgment, which is True in the _run_oneshot mode

nautilus_trader/nautilus_trader/backtest/engine.pyx

Lines 909 to 955 in 4535ca4

	def run(
	self,
	start: datetime | str | int | None = None,
	end: datetime | str | int | None = None,
	run_config_id: str | None = None,
	streaming: bool = False,
	) -> None:
	"""
	Run a backtest.
	At the end of the run the trader and strategies will be stopped, then
	post-run analysis performed.
	If more data than can fit in memory is to be run through the backtest
	engine, then `streaming` mode can be utilized. The expected sequence is as
	follows:
	- Add initial data batch and strategies.
	- Call `run(streaming=True)`.
	- Call `clear_data()`.
	- Add next batch of data stream.
	- Call either `run(streaming=False)` or `end()`. When there is no more data to run on.
	Parameters
	----------
	start : datetime or str or int, optional
	The start datetime (UTC) for the backtest run.
	If ``None`` engine runs from the start of the data.
	end : datetime or str or int, optional
	The end datetime (UTC) for the backtest run.
	If ``None`` engine runs to the end of the data.
	run_config_id : str, optional
	The tokenized `BacktestRunConfig` ID.
	streaming : bool, default False
	If running in streaming mode. If False, then will end the backtest
	following the run iterations.
	Raises
	------
	ValueError
	If no data has been added to the engine.
	ValueError
	If the `start` is >= the `end` datetime.
	"""
	self._run(start, end, run_config_id)
	if not streaming:
	self.end()

[ZzzzHeJ]

Comment out this code to make it work？

nautilus_trader/nautilus_trader/analysis/analyzer.py

Line 143 in b65c810

self._returns = pd.Series(dtype=float64)