def get_results(self): """ Return a dict with all important results & stats. """ # Equity equity_s = pd.Series(self.equity).sort_index() # Returns returns_s = equity_s.pct_change().fillna(0.0) # Rolling Annualised Sharpe rolling = returns_s.rolling(window=self.periods) rolling_sharpe_s = np.sqrt( self.periods) * (rolling.mean() / rolling.std()) # Cummulative Returns # YOE log(1+Rt)~Rt, por tanto exp(log(1+Rt)+log(1+Rt-1)+...))~Rt*Rt-1*... cum_returns_s = np.exp(np.log(1 + returns_s).cumsum()) # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(cum_returns_s) statistics = {} # Equity statistics statistics["sharpe"] = perf.create_sharpe_ratio( returns_s, self.periods) statistics["drawdowns"] = dd_s # TODO: need to have max_drawdown so it can be printed at end of test statistics["max_drawdown"] = max_dd statistics["max_drawdown_pct"] = max_dd statistics["max_drawdown_duration"] = dd_dur statistics["equity"] = equity_s statistics["returns"] = returns_s statistics["rolling_sharpe"] = rolling_sharpe_s statistics["cum_returns"] = cum_returns_s positions = self._get_positions() if positions is not None: statistics["positions"] = positions # Benchmark statistics if benchmark ticker specified if self.benchmark is not None: equity_b = pd.Series(self.equity_benchmark).sort_index() returns_b = equity_b.pct_change().fillna(0.0) rolling_b = returns_b.rolling(window=self.periods) rolling_sharpe_b = np.sqrt( self.periods) * (rolling_b.mean() / rolling_b.std()) cum_returns_b = np.exp(np.log(1 + returns_b).cumsum()) dd_b, max_dd_b, dd_dur_b = perf.create_drawdowns(cum_returns_b) statistics["sharpe_b"] = perf.create_sharpe_ratio(returns_b) statistics["drawdowns_b"] = dd_b statistics["max_drawdown_pct_b"] = max_dd_b statistics["max_drawdown_duration_b"] = dd_dur_b statistics["equity_b"] = equity_b statistics["returns_b"] = returns_b statistics["rolling_sharpe_b"] = rolling_sharpe_b statistics["cum_returns_b"] = cum_returns_b return statistics
def get_results(self): """ Return a dict with all important results & stats. """ # Equity equity_s = pd.Series(self.equity).sort_index() # Returns returns_s = equity_s.pct_change().fillna(0.0) # Rolling Annualised Sharpe rolling = returns_s.rolling(window=self.periods) rolling_sharpe_s = np.sqrt(self.periods) * ( rolling.mean() / rolling.std() ) # Cummulative Returns cum_returns_s = np.exp(np.log(1 + returns_s).cumsum()) # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(cum_returns_s) statistics = {} # Equity statistics statistics["sharpe"] = perf.create_sharpe_ratio( returns_s, self.periods ) statistics["drawdowns"] = dd_s # TODO: need to have max_drawdown so it can be printed at end of test statistics["max_drawdown"] = max_dd statistics["max_drawdown_pct"] = max_dd statistics["max_drawdown_duration"] = dd_dur statistics["equity"] = equity_s statistics["returns"] = returns_s statistics["rolling_sharpe"] = rolling_sharpe_s statistics["cum_returns"] = cum_returns_s statistics["positions"] = self._get_positions() # Benchmark statistics if benchmark ticker specified if self.benchmark is not None: equity_b = pd.Series(self.equity_benchmark).sort_index() returns_b = equity_b.pct_change().fillna(0.0) rolling_b = returns_b.rolling(window=self.periods) rolling_sharpe_b = np.sqrt(self.periods) * ( rolling_b.mean() / rolling_b.std() ) cum_returns_b = np.exp(np.log(1 + returns_b).cumsum()) dd_b, max_dd_b, dd_dur_b = perf.create_drawdowns(cum_returns_b) statistics["sharpe_b"] = perf.create_sharpe_ratio(returns_b) statistics["drawdowns_b"] = dd_b statistics["max_drawdown_pct_b"] = max_dd_b statistics["max_drawdown_duration_b"] = dd_dur_b statistics["equity_b"] = equity_b statistics["returns_b"] = returns_b statistics["rolling_sharpe_b"] = rolling_sharpe_b statistics["cum_returns_b"] = cum_returns_b return statistics
def get_results(self, equity_df): """ Return a dict with all important results & stats. """ # Returns equity_df["returns"] = equity_df["Equity"].pct_change().fillna(0.0) # Cummulative Returns equity_df["cum_returns"] = np.exp(np.log(1 + equity_df["returns"]).cumsum()) # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(equity_df["cum_returns"]) # Equity statistics statistics = {} statistics["sharpe"] = perf.create_sharpe_ratio( equity_df["returns"], self.periods ) statistics["drawdowns"] = dd_s statistics["max_drawdown"] = max_dd statistics["max_drawdown_pct"] = max_dd statistics["max_drawdown_duration"] = dd_dur statistics["equity"] = equity_df["Equity"] statistics["returns"] = equity_df["returns"] statistics["cum_returns"] = equity_df["cum_returns"] return statistics
def get_results(self): """ Return a dict with all important results & stats. """ # Equity equity_s = pd.Series(self.equity).sort_index() # Returns returns_s = equity_s.pct_change().fillna(0.0) # Cummulative Returns cum_returns_s = np.exp(np.log(1 + returns_s).cumsum()) # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(cum_returns_s) statistics = {} # Equity statistics statistics["sharpe"] = perf.create_sharpe_ratio( returns_s, self.periods) statistics["drawdowns"] = dd_s # TODO: need to have max_drawdown so it can be printed at end of test statistics["max_drawdown"] = max_dd statistics["max_drawdown_pct"] = max_dd statistics["max_drawdown_duration"] = dd_dur statistics["equity"] = equity_s statistics["returns"] = returns_s statistics["cum_returns"] = cum_returns_s statistics["positions"] = self._get_positions() # Benchmark statistics if benchmark ticker specified if self.benchmark is not None: equity_b = pd.Series(self.equity_benchmark).sort_index() returns_b = equity_b.pct_change().fillna(0.0) cum_returns_b = np.exp(np.log(1 + returns_b).cumsum()) dd_b, max_dd_b, dd_dur_b = perf.create_drawdowns(cum_returns_b) statistics["sharpe_b"] = perf.create_sharpe_ratio(returns_b) statistics["drawdowns_b"] = dd_b statistics["max_drawdown_pct_b"] = max_dd_b statistics["max_drawdown_duration_b"] = dd_dur_b statistics["equity_b"] = equity_b statistics["returns_b"] = returns_b statistics["cum_returns_b"] = cum_returns_b return statistics
def _calculate_statistics(self, curve): """ Creates a dictionary of various statistics associated with the backtest of a trading strategy via a supplied equity curve. All Pandas Series indexed by date-time are converted into milliseconds since epoch representation. Parameters ---------- curve : `pd.DataFrame` The equity curve DataFrame. Returns ------- `dict` The statistics dictionary. """ stats = {} # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(curve['CumReturns']) # Equity curve and returns stats['equity_curve'] = JSONStatistics._series_to_tuple_list(curve['Equity']) stats['returns'] = JSONStatistics._series_to_tuple_list(curve['Returns']) stats['cum_returns'] = JSONStatistics._series_to_tuple_list(curve['CumReturns']) # Month/year aggregated returns stats['monthly_agg_returns'] = self._calculate_monthly_aggregated_returns(curve['Returns']) stats['monthly_agg_returns_hc'] = self._calculate_monthly_aggregated_returns_hc(curve['Returns']) stats['yearly_agg_returns'] = self._calculate_yearly_aggregated_returns(curve['Returns']) stats['yearly_agg_returns_hc'] = self._calculate_yearly_aggregated_returns_hc(curve['Returns']) # Returns quantiles stats['returns_quantiles'] = self._calculate_returns_quantiles(curve['Returns']) stats['returns_quantiles_hc'] = self._calculate_returns_quantiles_hc(stats['returns_quantiles']) # Drawdown statistics stats['drawdowns'] = JSONStatistics._series_to_tuple_list(dd_s) stats['max_drawdown'] = max_dd stats['max_drawdown_duration'] = dd_dur # Performance stats['mean_returns'] = np.mean(curve['Returns']) stats['stdev_returns'] = np.std(curve['Returns']) stats['cagr'] = perf.create_cagr(curve['CumReturns'], self.periods) stats['annualised_vol'] = np.std(curve['Returns']) * np.sqrt(self.periods) stats['sharpe'] = perf.create_sharpe_ratio(curve['Returns'], self.periods) stats['sortino'] = perf.create_sortino_ratio(curve['Returns'], self.periods) return stats
def _plot_txt_curve(self, stats): """ Output the statistics for the equity curve """ returns = stats["returns"] cum_returns = stats['cum_returns'] if 'positions' not in stats: trd_yr = 0 else: positions = stats['positions'] trd_yr = positions.shape[0] / ( (returns.index[-1] - returns.index[0]).days / 365.0) tot_ret = cum_returns[-1] - 1.0 cagr = perf.create_cagr(cum_returns, self.periods) sharpe = perf.create_sharpe_ratio(returns, self.periods) sortino = perf.create_sortino_ratio(returns, self.periods) rsq = perf.rsquared(range(cum_returns.shape[0]), cum_returns) dd, dd_max, dd_dur = perf.create_drawdowns(cum_returns) header = ["Performance", "Value"] rows = [["Total Return", "{:.0%}".format(tot_ret)], ["CAGR", "{:.2%}".format(cagr)], ["Sharpe Ratio", "{:.2f}".format(sharpe)], ["Sortino Ratio", "{:.2f}".format(sortino)], [ "Annual Volatility", "{:.2%}".format(returns.std() * np.sqrt(252)) ], ["R-Squared", '{:.2f}'.format(rsq)], ["Max Daily Drawdown", '{:.2%}'.format(dd_max)], ["Max Drawdown Duration", '{:.0f}'.format(dd_dur)], ["Trades per Year", '{:.1f}'.format(trd_yr)]] table = (Table().add(header, rows).set_global_opts( title_opts=opts.ComponentTitleOpts(title="Curve"))) return table
def _plot_txt_curve(self, stats, ax=None, **kwargs): """ Outputs the statistics for the equity curve. """ def format_perc(x, pos): return '%.0f%%' % x returns = stats["returns"] cum_returns = stats['cum_returns'] if 'positions' not in stats: trd_yr = 0 else: positions = stats['positions'] trd_yr = positions.shape[0] / ( (returns.index[-1] - returns.index[0]).days / 365.0 ) if ax is None: ax = plt.gca() y_axis_formatter = FuncFormatter(format_perc) ax.yaxis.set_major_formatter(FuncFormatter(y_axis_formatter)) tot_ret = cum_returns[-1] - 1.0 cagr = perf.create_cagr(cum_returns, self.periods) sharpe = perf.create_sharpe_ratio(returns, self.periods) sortino = perf.create_sortino_ratio(returns, self.periods) rsq = perf.rsquared(range(cum_returns.shape[0]), cum_returns) dd, dd_max, dd_dur = perf.create_drawdowns(cum_returns) ax.text(0.25, 8.9, 'Total Return', fontsize=8) ax.text(7.50, 8.9, '{:.0%}'.format(tot_ret), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 7.9, 'CAGR', fontsize=8) ax.text(7.50, 7.9, '{:.2%}'.format(cagr), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 6.9, 'Sharpe Ratio', fontsize=8) ax.text(7.50, 6.9, '{:.2f}'.format(sharpe), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 5.9, 'Sortino Ratio', fontsize=8) ax.text(7.50, 5.9, '{:.2f}'.format(sortino), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 4.9, 'Annual Volatility', fontsize=8) ax.text(7.50, 4.9, '{:.2%}'.format(returns.std() * np.sqrt(252)), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 3.9, 'R-Squared', fontsize=8) ax.text(7.50, 3.9, '{:.2f}'.format(rsq), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 2.9, 'Max Daily Drawdown', fontsize=8) ax.text(7.50, 2.9, '{:.2%}'.format(dd_max), color='red', fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 1.9, 'Max Drawdown Duration', fontsize=8) ax.text(7.50, 1.9, '{:.0f}'.format(dd_dur), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 0.9, 'Trades per Year', fontsize=8) ax.text(7.50, 0.9, '{:.1f}'.format(trd_yr), fontweight='bold', horizontalalignment='right', fontsize=8) ax.set_title('Curve', fontweight='bold') if self.benchmark is not None: returns_b = stats['returns_b'] equity_b = stats['cum_returns_b'] tot_ret_b = equity_b[-1] - 1.0 cagr_b = perf.create_cagr(equity_b) sharpe_b = perf.create_sharpe_ratio(returns_b) sortino_b = perf.create_sortino_ratio(returns_b) rsq_b = perf.rsquared(range(equity_b.shape[0]), equity_b) dd_b, dd_max_b, dd_dur_b = perf.create_drawdowns(equity_b) ax.text(9.75, 8.9, '{:.0%}'.format(tot_ret_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 7.9, '{:.2%}'.format(cagr_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 6.9, '{:.2f}'.format(sharpe_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 5.9, '{:.2f}'.format(sortino_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 4.9, '{:.2%}'.format(returns_b.std() * np.sqrt(252)), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 3.9, '{:.2f}'.format(rsq_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 2.9, '{:.2%}'.format(dd_max_b), color='red', fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(9.75, 1.9, '{:.0f}'.format(dd_dur_b), fontweight='bold', horizontalalignment='right', fontsize=8) ax.set_title('Curve vs. Benchmark', fontweight='bold') ax.grid(False) ax.spines['top'].set_linewidth(2.0) ax.spines['bottom'].set_linewidth(2.0) ax.spines['right'].set_visible(False) ax.spines['left'].set_visible(False) ax.get_yaxis().set_visible(False) ax.get_xaxis().set_visible(False) ax.set_ylabel('') ax.set_xlabel('') ax.axis([0, 10, 0, 10]) return ax
def _plot_txt_curve(self, stats, bench_stats=None, ax=None, **kwargs): """ Outputs the statistics for the equity curve. """ def format_perc(x, pos): return '%.0f%%' % x if ax is None: ax = plt.gca() y_axis_formatter = FuncFormatter(format_perc) ax.yaxis.set_major_formatter(FuncFormatter(y_axis_formatter)) # Strategy statistics returns = stats["returns"] cum_returns = stats['cum_returns'] tot_ret = cum_returns[-1] - 1.0 cagr = perf.create_cagr(cum_returns, self.periods) sharpe = perf.create_sharpe_ratio(returns, self.periods) sortino = perf.create_sortino_ratio(returns, self.periods) dd, dd_max, dd_dur = perf.create_drawdowns(cum_returns) # Benchmark statistics if bench_stats is not None: bench_returns = bench_stats["returns"] bench_cum_returns = bench_stats['cum_returns'] bench_tot_ret = bench_cum_returns[-1] - 1.0 bench_cagr = perf.create_cagr(bench_cum_returns, self.periods) bench_sharpe = perf.create_sharpe_ratio(bench_returns, self.periods) bench_sortino = perf.create_sortino_ratio(bench_returns, self.periods) bench_dd, bench_dd_max, bench_dd_dur = perf.create_drawdowns(bench_cum_returns) # Strategy Values ax.text(7.50, 8.2, 'Strategy', fontweight='bold', horizontalalignment='right', fontsize=8, color='green') ax.text(0.25, 6.9, 'Total Return', fontsize=8) ax.text(7.50, 6.9, '{:.0%}'.format(tot_ret), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 5.9, 'CAGR', fontsize=8) ax.text(7.50, 5.9, '{:.2%}'.format(cagr), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 4.9, 'Sharpe Ratio', fontsize=8) ax.text(7.50, 4.9, '{:.2f}'.format(sharpe), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 3.9, 'Sortino Ratio', fontsize=8) ax.text(7.50, 3.9, '{:.2f}'.format(sortino), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 2.9, 'Annual Volatility', fontsize=8) ax.text(7.50, 2.9, '{:.2%}'.format(returns.std() * np.sqrt(252)), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 1.9, 'Max Daily Drawdown', fontsize=8) ax.text(7.50, 1.9, '{:.2%}'.format(dd_max), color='red', fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(0.25, 0.9, 'Max Drawdown Duration (Days)', fontsize=8) ax.text(7.50, 0.9, '{:.0f}'.format(dd_dur), fontweight='bold', horizontalalignment='right', fontsize=8) # Benchmark Values if bench_stats is not None: ax.text(10.0, 8.2, 'Benchmark', fontweight='bold', horizontalalignment='right', fontsize=8, color='gray') ax.text(10.0, 6.9, '{:.0%}'.format(bench_tot_ret), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 5.9, '{:.2%}'.format(bench_cagr), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 4.9, '{:.2f}'.format(bench_sharpe), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 3.9, '{:.2f}'.format(bench_sortino), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 2.9, '{:.2%}'.format(bench_returns.std() * np.sqrt(252)), fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 1.9, '{:.2%}'.format(bench_dd_max), color='red', fontweight='bold', horizontalalignment='right', fontsize=8) ax.text(10.0, 0.9, '{:.0f}'.format(bench_dd_dur), fontweight='bold', horizontalalignment='right', fontsize=8) ax.set_title('Equity Curve', fontweight='bold') ax.grid(False) ax.spines['top'].set_linewidth(2.0) ax.spines['bottom'].set_linewidth(2.0) ax.spines['right'].set_visible(False) ax.spines['left'].set_visible(False) ax.get_yaxis().set_visible(False) ax.get_xaxis().set_visible(False) ax.set_ylabel('') ax.set_xlabel('') ax.axis([0, 10, 0, 10]) return ax
def get_results(self): """ Return a dict with all important results & stats. """ # Equity equity_s = pd.Series(self.equity).sort_index() # Returns returns_s = equity_s.pct_change().fillna(0.0) # Rolling Annualised Sharpe rolling = returns_s.rolling(window=self.periods) rolling_sharpe_s = np.sqrt( self.periods) * (rolling.mean() / rolling.std()) # Cummulative Returns cum_returns_s = np.exp(np.log(1 + returns_s).cumsum()) # Drawdown, max drawdown, max drawdown duration dd_s, max_dd, dd_dur = perf.create_drawdowns(cum_returns_s) statistics = {} # Equity statistics statistics["sharpe"] = perf.create_sharpe_ratio( returns_s, self.periods) statistics["drawdowns"] = dd_s # TODO: need to have max_drawdown so it can be printed at end of test statistics["max_drawdown"] = max_dd statistics["max_drawdown_pct"] = max_dd statistics["max_drawdown_duration"] = dd_dur statistics["equity"] = equity_s statistics["returns"] = returns_s statistics["rolling_sharpe"] = rolling_sharpe_s statistics["cum_returns"] = cum_returns_s positions = self._get_positions() if positions is not None: statistics["positions"] = positions durations = (positions['end_timestamp'] - positions['start_timestamp']) / pd.Timedelta(days=1) avg_duration = durations.mean() statistics["max_duration"] = '{:.2f}'.format(durations.max()) statistics["min_duration"] = '{:.2f}'.format(durations.min()) statistics["std_duration"] = '{:.2f}'.format(durations.std()) statistics["avg_duration"] = '{:.2f}'.format(avg_duration) # Benchmark statistics if benchmark ticker specified if self.benchmark is not None: equity_b = pd.Series(self.equity_benchmark).sort_index() returns_b = equity_b.pct_change().fillna(0.0) rolling_b = returns_b.rolling(window=self.periods) rolling_sharpe_b = np.sqrt( self.periods) * (rolling_b.mean() / rolling_b.std()) cum_returns_b = np.exp(np.log(1 + returns_b).cumsum()) dd_b, max_dd_b, dd_dur_b = perf.create_drawdowns(cum_returns_b) statistics["sharpe_b"] = perf.create_sharpe_ratio(returns_b) statistics["drawdowns_b"] = dd_b statistics["max_drawdown_pct_b"] = max_dd_b statistics["max_drawdown_duration_b"] = dd_dur_b statistics["equity_b"] = equity_b statistics["returns_b"] = returns_b statistics["rolling_sharpe_b"] = rolling_sharpe_b statistics["cum_returns_b"] = cum_returns_b return statistics