所以我有下面的循环，但想知道是否有任何方法可以重写我的任何代码，使其运行速度明显更快（目前运行大约需要30分钟）。任何建议都表示赞赏，因为我仍然是编程和熊猫的初学者。谢谢！
all_dates是datetime.dates的列表
loan_matches_by_month_36是一个字典，其中的键是datetime.dates，值是该日期发放的贷款的数据帧。
aggregate_cashflow_dataframe_platform = pd.DataFrame([])
investment_per_loan = 25
for month in all_dates:
if month <= max_done_date:
for loan in loan_matches_by_month_36[month].index:
p_n = loan_matches_by_month_36[month].ix[loan,'p_n']
fraction = investment_per_loan/(loan_matches_by_month_36[month].ix[loan,'funded'])
p_adjust = loan_matches_by_month_36[month].ix[loan,'p_adjust']*fraction
installment = loan_matches_by_month_36[month].ix[loan,'installment']*fraction
cashflow_vector = {loan: [-investment_per_loan] + [installment] * (p_n-1) +
[installment + p_adjust]}
cashflow_dataframe = pd.DataFrame(cashflow_vector)
columns = pd.date_range(start = month, end = (month + relativedelta(
months=+len(cashflow_dataframe))), freq = 'M').shift(15, freq=pd.datetools.day)
cashflow_dataframe = cashflow_dataframe.T
cashflow_dataframe.columns = columns
cashflow_dataframe['month'] = month
cashflow_dataframe.set_index('month', append = True, inplace = True)
cashflow_dataframe.index.names = ['index', 'month']
result = aggregate_cashflow_dataframe_platform.append(cashflow_dataframe)
aggregate_cashflow_dataframe_platform = result
print month