Conditional short-term trend signals #
Get packages and JPMaQS data #
import os
import numpy as np
import pandas as pd
import itertools
import macrosynergy.management as msm
import macrosynergy.panel as msp
import macrosynergy.pnl as msn
import macrosynergy.signal as mss
import macrosynergy.learning as msl
import macrosynergy.visuals as msv
from macrosynergy.download import JPMaQSDownload
from macrosynergy.management.utils import merge_categories
import warnings
warnings.simplefilter("ignore")
# Cross-sections of currencies
cids_ecos = ["EUR", "USD"]
# Cross-sections of commodities
cids_nfm = ["GLD", "SIV", "PAL", "PLT"]
cids_fme = ["ALM", "CPR", "LED", "NIC", "TIN", "ZNC"]
cids_ene = ["BRT", "WTI", "NGS", "GSO", "HOL"]
cids_sta = ["COR", "WHT", "SOY", "CTN"]
cids_liv = ["CAT", "HOG"]
cids_mis = ["CFE", "SGR", "NJO", "CLB"]
cids_com = cids_nfm + cids_fme + cids_ene + cids_sta + cids_liv + cids_mis
# Mixed contracts identifiers
cids_alc = sorted(cids_ecos) + sorted(cids_com) # all contracts list
# Country categories
# CPI inflation
cpi = [
"CPIH_SA_P1M1ML12",
"CPIC_SA_P1M1ML12",
"CPIC_SJA_P6M6ML6AR",
"INFE2Y_JA",
]
# PPI inflation
ppi = [
"PGDPTECH_SA_P1M1ML12_3MMA",
"PGDPTECHX_SA_P1M1ML12_3MMA",
"PPIH_NSA_P1M1ML12_3MMA",
"PPIH_SA_P6M6ML6AR",
]
# Other inflation-related indicators
opi = [
"WAGES_NSA_P1M1ML12_3MMA",
"WAGES_NSA_P1Q1QL4",
"NRSALES_SA_P1M1ML12_3MMA",
"PCREDITBN_SJA_P1M1ML12",
"HPI_SA_P1M1ML12_3MMA",
"HPI_SA_P1Q1QL4",
]
# Benchmarks
bms = ["RGDP_SA_P1Q1QL4_20QMM", "INFTARGET_NSA", "WFORCE_NSA_P1Y1YL1_5YMM"]
# Equity returns
eqr = ["EQXR_NSA", "EQXR_VT10"]
ecos = cpi + ppi + opi + bms + eqr
# Commodity categories
com = ["COXR_NSA", "COXR_VT10"]
# Tickers
tix_eco = [cid + "_" + xcat for cid in cids_ecos for xcat in ecos]
tix_com = [cid + "_" + xcat for cid in cids_com for xcat in com]
tickers = tix_eco + tix_com
# Download series from J.P. Morgan DataQuery by tickers
start_date = "1990-01-01"
end_date = None
# Retrieve credentials
oauth_id = os.getenv("DQ_CLIENT_ID") # Replace with own client ID
oauth_secret = os.getenv("DQ_CLIENT_SECRET") # Replace with own secret
# Download from DataQuery
downloader = JPMaQSDownload(client_id=oauth_id, client_secret=oauth_secret)
df = downloader.download(
tickers=tickers,
start_date=start_date,
end_date=end_date,
metrics=["value"],
suppress_warning=True,
show_progress=True,
)
dfx = df.copy()
dfx.info()
Downloading data from JPMaQS.
Timestamp UTC: 2025-03-05 16:03:50
Connection successful!
Requesting data: 100%|██████████| 5/5 [00:01<00:00, 4.93it/s]
Downloading data: 100%|██████████| 5/5 [00:30<00:00, 6.05s/it]
Some expressions are missing from the downloaded data. Check logger output for complete list.
4 out of 88 expressions are missing. To download the catalogue of all available expressions and filter the unavailable expressions, set `get_catalogue=True` in the call to `JPMaQSDownload.download()`.
Some dates are missing from the downloaded data.
2 out of 9180 dates are missing.
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 648525 entries, 0 to 648524
Data columns (total 4 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 real_date 648525 non-null datetime64[ns]
1 cid 648525 non-null object
2 xcat 648525 non-null object
3 value 648525 non-null float64
dtypes: datetime64[ns](1), float64(1), object(2)
memory usage: 19.8+ MB
Availability and blacklisting #
Rename #
Rename quarterly tickers to roughly equivalent monthly tickers to simplify subsequent operations.
dict_repl = {
"WAGES_NSA_P1Q1QL4": "WAGES_NSA_P1M1ML12_3MMA",
"HPI_SA_P1Q1QL4": "HPI_SA_P1M1ML12_3MMA",
}
for key, value in dict_repl.items():
dfx["xcat"] = dfx["xcat"].str.replace(key, value)
Backfill inflation target #
# Backward-extension of inflation target with oldest available
# Duplicate targets
cidx = cids_ecos
calcs = [f"INFTARGET_BX = INFTARGET_NSA"]
dfa = msp.panel_calculator(dfx, calcs, cids=cidx)
# Add all dates back to 1990 to the frame, filling "value " with NaN
all_dates = np.sort(dfx['real_date'].unique())
all_combinations = pd.DataFrame(
list(itertools.product(dfa['cid'].unique(), dfa['xcat'].unique(), all_dates)),
columns=['cid', 'xcat', 'real_date']
)
dfax = pd.merge(all_combinations, dfa, on=['cid', 'xcat', 'real_date'], how='left')
# Backfill the values with first target value
dfax = dfax.sort_values(by=['cid', 'xcat', 'real_date'])
dfax['value'] = dfax.groupby(['cid', 'xcat'])['value'].bfill()
dfx = msm.update_df(dfx, dfax)
# Extended effective inflation target by hierarchical merging
hierarchy = ["INFTARGET_NSA", "INFTARGET_BX"]
dfa = merge_categories(dfx, xcats=hierarchy, new_xcat="INFTARGETX_NSA")
dfx = msm.update_df(dfx, dfa)
Check availability #
xcatx = ecos
cidx = cids_ecos
msm.check_availability(df=dfx, xcats=xcatx, cids=cidx, missing_recent=False)
xcatx = com
cidx = cids_com
msm.check_availability(df=dfx, xcats=xcatx, cids=cidx, missing_recent=False)
Factor calculation #
Short-term return trends #
dict_xrs = {"COXR_NSA": cids_com, "EQXR_NSA": cids_ecos}
dfa = pd.DataFrame(columns=dfx.columns)
for xr, cidx in dict_xrs.items():
calcs = [
f"XRI = ( {xr} ).cumsum()",
f"XRI_3DMA = XRI.rolling(3).mean()",
f"XRI_3DMA = XRI.rolling(3).mean()",
f"XRI_5DMA = XRI.rolling(5).mean()",
f"XRI_10DMA = XRI.rolling(10).mean()",
f"XRI_1Dv5D = ( XRI - XRI_5DMA )",
f"XRI_1Dv3D = ( XRI - XRI_3DMA )",
f"XRI_3Dv10D = ( XRI_3DMA - XRI_10DMA )",
]
dfaa = msp.panel_calculator(dfx, calcs, cids=cidx)
dfa = msm.update_df(dfa, dfaa)
dfx = msm.update_df(dfx, dfa)
trends = ["XRI_3Dv10D", "XRI_1Dv5D", "XRI_1Dv3D"]
xcatx = ["XRI_3DMA", "XRI_10DMA"]
cidx = ["WTI"]
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=1,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.5,
)
xcatx = trends
cidx = cids_alc
sx = "1995-01-01"
msp.view_ranges(
dfx,
xcats=xcatx,
cids=cidx,
kind="bar",
start=sx,
sort_cids_by="mean", # here sorted by standard deviations
title=None,
ylab=None,
xcat_labels=None,
size=(12, 6),
)
xcatx = trends
cidx = cids_alc
sx = "2024-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=4,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.5,
)
xcatx = trends
cid_groups = {
"FOD": {"csts": cids_sta + cids_liv + ["CFE", "SGR", "NJO"], "wgts": None},
"ENY": {"csts": cids_ene, "wgts": None},
"FON": {"csts": ["FOD", "ENY"], "wgts": (2, 1)},
"MTS": {"csts": cids_fme + cids_nfm, "wgts": None},
"COM": {"csts": ["FOD", "ENY", "MTS"], "wgts": None},
"EQY": {"csts": ["EUR", "USD"], "wgts": None},
"ACE": {"csts": ["COM", "EQY"], "wgts": None},
}
for group, value in cid_groups.items():
for xc in xcatx:
dfa = msp.linear_composite(
df=dfx,
xcats=xc,
cids=value["csts"],
weights=value["wgts"],
complete_cids=False,
new_cid=group,
)
dfx = msm.update_df(dfx, dfa)
gcids = list(cid_groups.keys())
secs =["FOD", "ENY", "MTS", "COM", "EQY", "ACE"] # selected sectors
xcatx = "XRI_3Dv10D"
cidx = secs
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cid_labels=["Agriculture", "Energy", "Metals", "Commodities", "Equities", "All"],
cids=cidx,
ncol=3,
start=sx,
title="Short-term commodity and equity futures trends (3 days vs 10 days moving averages)",
title_fontsize=20,
same_y=True,
size = (12, 7),
aspect = 1.5,
)
Conditioning factors #
cidx = cids_ecos
calcs = []
# Excess inflation measures
xcatx = cpi + ppi + ["HPI_SA_P1M1ML12_3MMA"]
for xc in xcatx:
calcs.append(f"X{xc} = {xc} - INFTARGETX_NSA ")
# Excess wage growth
xcatx = ["WAGES_NSA_P1M1ML12_3MMA"]
for xc in xcatx:
calcs.append(f"X{xc} = {xc} - RGDP_SA_P1Q1QL4_20QMM - INFTARGETX_NSA + WFORCE_NSA_P1Y1YL1_5YMM ")
# Excess sales and credit growth
xcatx = ["NRSALES_SA_P1M1ML12_3MMA", "PCREDITBN_SJA_P1M1ML12"]
for xc in xcatx:
calcs.append(f"X{xc} = {xc} - RGDP_SA_P1Q1QL4_20QMM - INFTARGETX_NSA")
# Actual calculation and addition
dfa = msp.panel_calculator(dfx, calcs, cids=cidx)
dfx = msm.update_df(dfx, dfa)
# Lists of relevant excess indicators
xall = [s.split(' ', 1)[0] for s in calcs]
xcpi = [x for x in xall if "CPI" in x or "INFE" in x]
xppi = [x for x in xall if "PPI" in x or "PGDPTECH" in x]
xopi = [x for x in xall if x not in xcpi + xppi]
xinf = xcpi + xppi + xopi
xopi
['XHPI_SA_P1M1ML12_3MMA',
'XWAGES_NSA_P1M1ML12_3MMA',
'XNRSALES_SA_P1M1ML12_3MMA',
'XPCREDITBN_SJA_P1M1ML12']
xcatx = xcpi
cidx = cids_ecos
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.7,
)
xcatx = xppi
cidx = cids_ecos
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.7,
)
xcatx = xopi
cidx = cids_ecos
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.7,
)
# Normalization of broadest excess inflation
xcatx = xinf
cidx = cids_ecos
for xc in xcatx:
dfa = msp.make_zn_scores(
dfx,
xcat=xc,
cids=cidx,
neutral="zero",
thresh=2,
est_freq="M",
pan_weight=1,
postfix="_ZN",
)
dfx = msm.update_df(dfx, dfa)
# Lists of normalized excess indicators
xcpiz = [s + "_ZN" for s in xcpi]
xppiz = [s + "_ZN" for s in xppi]
xopiz = [s + "_ZN" for s in xopi]
xinfz = [s + "_ZN" for s in xinf]
xcatx = xinfz
cidx = cids_ecos
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.7,
)
# Groupwise linear combination
cidx = cids_ecos
xcat_groups = {
"XCPI": xcpiz,
"XPPI": xppiz,
"XOPI": xopiz,
}
for group, xcatx in xcat_groups.items():
dfa = msp.linear_composite(
df=dfx,
xcats=xcatx,
cids=cidx,
complete_xcats=False,
new_xcat=group,
)
dfx = msm.update_df(dfx, dfa)
# Re-scoring
comps = list(xcat_groups.keys())
for xc in comps:
dfa = msp.make_zn_scores(
dfx,
xcat=xc,
cids=cidx,
neutral="zero",
thresh=2,
est_freq="M",
pan_weight=1,
postfix="_ZN",
)
dfx = msm.update_df(dfx, dfa)
compz = [s + "_ZN" for s in comps]
# Final composite and re-normalization
cidx = cids_ecos
xcatx = compz
dfa = msp.linear_composite(
df=dfx,
xcats=xcatx,
cids=cidx,
complete_xcats=False,
new_xcat="XINFS",
)
dfx = msm.update_df(dfx, dfa)
dfa = msp.make_zn_scores(
dfx,
xcat="XINFS",
cids=cidx,
neutral="zero",
thresh=2,
est_freq="M",
pan_weight=1,
postfix="_ZN",
)
dfx = msm.update_df(dfx, dfa)
xcatx = compz + ["XINFS_ZN"]
cidx = "USD"
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
xcat_labels=[
"Excess CPI inflation score",
"Excess PPI inflation score",
"Excess credit, sales and house inflation score",
"Composite excess inflation score",
],
ncol=2,
start=sx,
title="U.S. economy: conditioning point-in-time factors for futures market trends",
title_fontsize=20,
xcat_grid=True,
same_y=True,
size = (10, 6),
aspect = 2,
)
xcatx = compz + ["XINFS_ZN"]
cidx = cids_ecos
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start=sx,
title=None,
same_y=True,
size = (12, 7),
aspect = 1.7,
)
Conditional short-term trends #
# Sign-conditioned trends
cidx = gcids
us_condz_s = ["USD_" + s for s in ['XCPI_ZN', 'XINFS_ZN']]
calcs = []
for trend in trends:
for cond in us_condz_s:
calcs += [f"{trend}_{cond[:-3]} = -1 * i{cond} * {trend}"]
dfa = msp.panel_calculator(dfx, calcs=calcs, cids=cidx)
dfx = msm.update_df(dfx, dfa)
xcatx = ["XRI_3Dv10D", "XRI_3Dv10D_USD_XINFS"]
cidx = secs
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
# cid_labels=["Agriculture", "Energy", "Metals", "Commodities", "Equities", "All"],
cids=cidx,
ncol=3,
start=sx,
title=None,
title_fontsize=20,
same_y=True,
size = (12, 7),
aspect = 1.5,
)
# Renaming to currency cids
mofs = ["_USD_XCPI", "_USD_XINFS", ""]
sm_trends = [trend + mof for trend in trends for mof in mofs]
cidx = cids_ecos
contracts = [
cid + "_" + smt for cid in gcids for smt in sm_trends
]
calcs = []
for contract in contracts:
calcs.append(f"{contract} = i{contract}")
dfa = msp.panel_calculator(dfx, calcs=calcs, cids=cidx)
dfx = msm.update_df(dfx, dfa)
# Negative of simple trends
cidx = cids_ecos
contracts = [
cid + "_" + smt for cid in gcids for smt in trends
]
calcs = []
for contract in contracts:
calcs.append(f"{contract}N = -1 * i{contract}")
dfa = msp.panel_calculator(dfx, calcs=calcs, cids=cidx)
dfx = msm.update_df(dfx, dfa)
xcatx = ["ACE_XRI_3Dv10DN", "ACE_XRI_3Dv10D_USD_XINFS"]
cidx = ["USD"]
sx = "1995-01-01"
msp.view_timelines(
dfx,
xcats=xcatx,
xcat_labels=["Unconditional futures trend (negative)", "Conditional futures trend"],
cids=cidx,
start=sx,
title="Unconditional and conditional composite futures return trends (conditional on composite excess inflation)",
title_fontsize=16,
same_y=True,
size = (16, 6),
aspect = 1.5,
)
Combined signals #
# Combine by trends and modifying factors across contracts
mofns = ["_USD_XCPI", "_USD_XINFS", "N"]
secs =["FOD", "ENY", "MTS", "COM", "EQY", "ACE"]
for trend in trends:
dfa = msp.linear_composite(
df=dfx,
xcats=[sec + "_" + trend + mof for mof in mofns for sec in secs],
cids=["USD"],
complete_xcats=False,
new_xcat=trend + "_USD_ALL_CM",
)
dfx = msm.update_df(dfx, dfa)
dfa = msp.linear_composite(
df=dfx,
xcats=[sec + "_" + trend + "N" for sec in secs],
cids=["USD"],
complete_xcats=False,
new_xcat=trend + "_USD_ALL_N",
)
dfx = msm.update_df(dfx, dfa)
Target returns #
cidx = cids_ecos
xcatx = ["EQXR_NSA", "EQXR_VT10"]
msp.view_timelines(
dfx,
xcats=xcatx,
cids=cidx,
ncol=2,
start="1995-01-01",
same_y=True,
cumsum=True,
)
Value checks #
All contracts and modifiers #
Specs and test #
trendx = ["XRI_3Dv10D", "XRI_1Dv5D"]
all_mods = ["_USD_ALL_N", "_USD_ALL_CM"]
all_cm_trends = [trend + mod for trend in trendx for mod in all_mods]
cidx = ["USD"]
dict_all_cmw = {
"sigs": all_cm_trends,
"targs": ["EQXR_NSA", "EQXR_VT10"],
"cidx": cidx,
"start": "1995-01-01",
"srr": None,
"pnls": None,
"pnls_lb": None,
}
dix = dict_all_cmw
sigs = dix["sigs"]
targ = dix["targs"][0] # assuming just one target
cidx = dix["cidx"]
start = dix["start"]
# Initialize the dictionary to store CategoryRelations instances
dict_cr = {}
for sig in sigs:
dict_cr[sig] = msp.CategoryRelations(
dfx,
xcats=[sig, targ],
cids=cidx,
freq="W",
lag=1,
xcat_aggs=["last", "sum"],
start=start,
)
# Plotting the results
crs = list(dict_cr.values())
crs_keys = list(dict_cr.keys())
msv.multiple_reg_scatter(
cat_rels=crs,
title="Futures trend signals and subsequent weekly S&P500 returns, 1995-2024",
ylab=None,
ncol=2,
nrow=2,
figsize=(17, 10),
prob_est="pool",
coef_box="lower right",
subplot_titles=[
"3d/10d futures trend, unconditional",
"3d/10d futures trend, conditional on macro",
"1d/5d futures trend, unconditional",
"1d/5d futures trend, conditional on macro",
],
)
Accuracy and correlation check #
dix = dict_all_cmw
sigx = dix["sigs"]
targx = dix["targs"][0]
cidx = dix["cidx"]
start = dix["start"]
srr = mss.SignalReturnRelations(
dfx,
cids=cidx,
sigs=sigx,
rets=targx,
freqs="W",
slip=0,
start=start,
)
dix["srr"] = srr
dix = dict_all_cmw
srr = dix["srr"]
colx = [
"accuracy",
"bal_accuracy",
"pos_sigr",
"pos_retr",
"pearson",
"pearson_pval",
"kendall",
"kendall_pval",
]
kills = ["Return", "Frequency", "Aggregation"]
display(
srr.multiple_relations_table()
.reset_index(level=kills, drop=True)
.astype("float")
.round(3)[colx]
)
dict_labels = {
"XRI_3Dv10D_USD_ALL_N": "3d/10d trend, unconditional",
"XRI_3Dv10D_USD_ALL_CM": "3d/10d trend, conditional on macro",
"XRI_1Dv5D_USD_ALL_N": "1d/5d trend, unconditional",
"XRI_1Dv5D_USD_ALL_CM": "1d/5d trend, conditional on macro",
}
srr.correlation_bars(
type="signals",
title="Linear correlation of commodity trend signals and next week's S&P500 returns",
size=(18, 7),
x_labels=dict_labels,
)
srr.accuracy_bars(
type="signals",
title="Accuracy ratios of commodity trend signals for next week's S&P500 returns",
size=(18, 7),
x_labels=dict_labels,
)
| accuracy | bal_accuracy | pos_sigr | pos_retr | pearson | pearson_pval | kendall | kendall_pval | |
|---|---|---|---|---|---|---|---|---|
| Signal | ||||||||
| XRI_1Dv5D_USD_ALL_CM | 0.522 | 0.524 | 0.480 | 0.57 | 0.058 | 0.023 | 0.046 | 0.006 |
| XRI_1Dv5D_USD_ALL_N | 0.499 | 0.509 | 0.428 | 0.57 | 0.031 | 0.221 | 0.029 | 0.089 |
| XRI_3Dv10D_USD_ALL_CM | 0.518 | 0.520 | 0.483 | 0.57 | 0.072 | 0.004 | 0.060 | 0.000 |
| XRI_3Dv10D_USD_ALL_N | 0.489 | 0.498 | 0.429 | 0.57 | 0.064 | 0.011 | 0.043 | 0.011 |
Naive PnL #
dix = dict_all_cmw
sigx = dix["sigs"]
targ = dix["targs"][1]
cidx = dix["cidx"]
start = dix["start"]
naive_pnl = msn.NaivePnL(
dfx,
ret=targ,
sigs=sigx,
cids=cidx,
start=start,
bms=["USD_EQXR_NSA"],
)
for sig in sigx:
naive_pnl.make_pnl(
sig,
sig_neg=False,
sig_op="zn_score_cs",
sig_add=0,
thresh=4,
rebal_freq="weekly",
vol_scale=10,
rebal_slip=0,
pnl_name=sig+"_ZN"
)
naive_pnl.make_long_pnl(vol_scale=10, label="Long only")
dix["pnls"] = naive_pnl
dix = dict_all_cmw
start = dix["start"]
sigx = ["XRI_3Dv10D_USD_ALL_N", "XRI_3Dv10D_USD_ALL_CM"]
naive_pnl = dix["pnls"]
pnls = [s + "_ZN" for s in sigx]
desc = [
"based on unconditional futures trends and vol-targeting",
"based on conditional futures trends and vol targeting",
]
labels = {key: desc for key, desc in zip(pnls, desc)}
naive_pnl.plot_pnls(
pnl_cats=pnls,
start=start,
title="S&P 500: Naive PnL of balanced long-short futures strategy",
title_fontsize=14,
xcat_labels=labels,
figsize=(12, 6),
)
df_eval = naive_pnl.evaluate_pnls(
pnl_cats=pnls,
start=start,
)
display(df_eval.transpose().astype("float").round(3))
| Return % | St. Dev. % | Sharpe Ratio | Sortino Ratio | Max 21-Day Draw % | Max 6-Month Draw % | Peak to Trough Draw % | Top 5% Monthly PnL Share | USD_EQXR_NSA correl | Traded Months | |
|---|---|---|---|---|---|---|---|---|---|---|
| xcat | ||||||||||
| XRI_3Dv10D_USD_ALL_N_ZN | 3.449 | 10.0 | 0.345 | 0.504 | -11.680 | -15.171 | -24.676 | 1.044 | 0.209 | 363.0 |
| XRI_3Dv10D_USD_ALL_CM_ZN | 4.539 | 10.0 | 0.454 | 0.689 | -8.473 | -17.633 | -19.908 | 0.951 | 0.047 | 363.0 |
dix = dict_all_cmw
sigx = dix["sigs"]
targ = dix["targs"][1]
cidx = dix["cidx"]
start = dix["start"]
naive_pnl = msn.NaivePnL(
dfx,
ret=targ,
sigs=sigx,
cids=cidx,
start=start,
bms=["USD_EQXR_NSA"],
)
for sig in sigx:
naive_pnl.make_pnl(
sig,
sig_neg=False,
sig_op="zn_score_cs",
sig_add=2,
thresh=2,
rebal_freq="weekly",
vol_scale=10,
rebal_slip=0,
pnl_name=sig+"_ZN"
)
naive_pnl.make_long_pnl(vol_scale=10, label="Long only")
dix["pnls_lb"] = naive_pnl
dix = dict_all_cmw
start = dix["start"]
sigx = ["XRI_3Dv10D_USD_ALL_CM"]
naive_pnl = dix["pnls_lb"]
pnls = ["Long only"] + [s + "_ZN" for s in sigx]
desc = [
"simple volatility-targeted S&P500 position",
"enhanced by considering conditional futures trends",
]
labels = {key: desc for key, desc in zip(pnls, desc)}
naive_pnl.plot_pnls(
pnl_cats=pnls,
start=start,
title="S&P 500 future, managed leveraged long strategies",
title_fontsize=14,
xcat_labels=labels,
figsize=(12, 6),
)
df_eval = naive_pnl.evaluate_pnls(
pnl_cats=pnls,
start=start,
)
display(df_eval.transpose().astype("float").round(3))
| Return % | St. Dev. % | Sharpe Ratio | Sortino Ratio | Max 21-Day Draw % | Max 6-Month Draw % | Peak to Trough Draw % | Top 5% Monthly PnL Share | USD_EQXR_NSA correl | Traded Months | |
|---|---|---|---|---|---|---|---|---|---|---|
| xcat | ||||||||||
| XRI_3Dv10D_USD_ALL_CM_ZN | 7.429 | 10.0 | 0.743 | 1.038 | -18.014 | -14.603 | -30.553 | 0.425 | 0.782 | 363.0 |
| Long only | 5.911 | 10.0 | 0.591 | 0.813 | -16.183 | -15.031 | -34.008 | 0.486 | 0.865 | 363.0 |
