cvxpy two:
time
When you invest in stocks you want to get a high return, but a low risk. There's also going to be a few constraints. That sounds like a job for cvxpy.
Notes
The code is not listed in the video, but if you want to repeat this yourself, you can use the following code to make the same charts.
Data Prep
from tqdm import tqdm
def make_dataf_generator(dataf, window=90, step=5):
i = 0
while i*step+window < len(dataf):
subset = dataf.iloc[0 + i*step:window+i*step]
yield subset.index[-1], subset
i += 1
df = pd.read_csv("stocks.csv").assign(Date = lambda d: pd.to_datetime(d['Date']))
subset_df = (df
.loc[lambda d: d['Date'] < pd.to_datetime('2010-01-01')]
.set_index('Date'))
groups = make_dataf_generator(subset_df)
x_vals = []
profit = []
timestamps = []
for ts, g in tqdm(groups):
# this is where you can change the risk preference
p, xs = solve_problem(g, risk_pref=1.5)
timestamps.append(ts)
x_vals.append(xs)
profit.append(p)
Chart One
plt.figure(figsize=(12, 3))
plt.plot(timestamps, profit);
Chart Two
plt.figure(figsize=(12, 3))
x_vals = np.array(x_vals)
for idx, stock in enumerate(tickers):
plt.plot(timestamps, x_vals[:, idx], label=stock)
plt.legend();
Feedback? See an issue? Something unclear? Feel free to mention it here.
If you want to be kept up to date, consider signing up for the newsletter.