Original paper
Abstract
We find that the stocks in which active mutual fund or hedge fund managers display the most conviction towards ex-ante, their “Best ideas,” outperform the market, as well as the other stocks in those managers' portfolios, by approximately 2.8 to 4.5 percent per year, depending on the benchmark employed. The vast majority of the other stocks managers hold do not exhibit significant outperformance. Thus, the organization of the money management industry appears to make it optimal for managers to introduce stocks into their portfolio that are not outperformers. We argue that investors would benefit if managers held more concentrated portfolios.
Keywords: mutual funds, managerial skill, market efficiency
Trading rules
- Create a group of mutual fund managers actively trading.
- Employ 13F submissions to identify primary stock selections of each manager.
- Allocate funds to stocks predominantly preferred as "top choices" by most managers.
Python code
Backtrader
Here’s the basic outline of the Backtrader python code for the given trading rules:
import backtrader as bt
import backtrader.feeds as btfeeds
import pandas as pd
class MutualFundStrategy(bt.Strategy):
def __init__(self):
self.selection_pool = self._establish_selection_pool()
def next(self):
best_ideas = self._get_best_ideas()
self._invest_in_best_ideas(best_ideas)
def _establish_selection_pool(self):
# Your code to establish a selection pool of active mutual fund managers
pass
def _utilize_13F_filings(self, manager):
# Your code to utilize 13F filings to pinpoint top stock picks for each manager
pass
def _get_best_ideas(self):
best_ideas = []
for manager in self.selection_pool:
top_picks = self._utilize_13F_filings(manager)
best_ideas.extend(top_picks)
# Your code to find stocks that are favored by the majority of managers as their "best ideas"
return best_ideas
def _invest_in_best_ideas(self, best_ideas):
# Your code to invest in the best ideas
pass
# Data feed setup and backtesting code
if __name__ == '__main__':
cerebro = bt.Cerebro()
data = btfeeds.GenericCSVData(
dataname='your-data-file.csv',
dtformat=('%Y-%m-%d'),
openinterest=-1
)
cerebro.adddata(data)
cerebro.addstrategy(MutualFundStrategy)
cerebro.run()Please note that this is just a basic outline of the code, and you will need to implement the logic for each function and provide the required data to make it functional.