Fix time rebalance

Fix Time Rebalance Examples

These examples show how to rebalance a portfolio on a fixed schedule — monthly, quarterly, yearly, or a custom schedule built with Or.

Monthly Rebalance with Fixed Ratio Allocation

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]
target_ratio = {"QQQ": 0.7, "BIL": 0.2, "GLD": 0.1}

data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'monthly_rebalance',
    [
        ti.Signal.Monthly(),              # trigger: end of month
        ti.Select.All(),                    # select all tickers
        ti.Weigh.Ratio(weights=target_ratio),  # fixed target weights
        ti.Action.Rebalance(),                    # execute trades
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

Monthly (Mid-Month) Rebalance with Equal Weighting

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]

data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'monthly_rebalance_mid',
    [
        ti.Signal.Monthly(day=15, closest_trading_day=True),  # 15th or next trading day
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

Quarterly Rebalance

The built-in shortcut:

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]

data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'quarterly_rebalance',
    [
        ti.Signal.Quarterly(months=[2, 5, 8, 11]),  # end of Feb/May/Aug/Nov
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

Note: Months [2, 5, 8, 11] are offset one month ahead of calendar quarter-ends (Mar/Jun/Sep/Dec) to avoid rebalancing on the same dates as most institutional investors, reducing market impact.

Or built manually with ti.Or when you need per-month customisation:

portfolio = ti.Portfolio(
    'quarterly_rebalance',
    [
        ti.Or(
            ti.Signal.Schedule(month=2),
            ti.Signal.Schedule(month=5),
            ti.Signal.Schedule(month=8),
            ti.Signal.Schedule(month=11),
        ),
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

Every 6 Months Rebalance

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]

data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'half_year_rebalance',
    [
        ti.Or(
            ti.Signal.Schedule(day=15, closest_trading_day=True, month=2),
            ti.Signal.Schedule(day=15, closest_trading_day=True, month=8),
        ),
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

Yearly Rebalance

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]

data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'yearly_rebalance',
    [
        ti.Signal.Schedule(day=15, closest_trading_day=True, month=7),
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

Branching: Or, And, Not

The algo stack runs top-to-bottom. Each algo can return True (continue) or False (stop). Branch operators let you compose conditions across multiple algos in a single stack position.

Operator	Behaviour
`ti.Or(a, b)`	Passes if any inner algo passes
`ti.And(a, b)`	Passes only if all inner algos pass
`ti.Not(a)`	Passes if the inner algo does not pass

`ti.And` — require multiple conditions simultaneously

Rebalance only at month-end and only if the current month is not December (year-end tax harvest window):

import tiportfolio as ti

tickers = ["QQQ", "BIL", "GLD"]
data = ti.fetch_data(tickers, start="2019-01-01", end="2024-12-31")

portfolio = ti.Portfolio(
    'monthly_skip_december',
    [
        ti.And(
            ti.Signal.Monthly(),
            ti.Not(ti.Signal.Schedule(month=12)),
        ),
        ti.Select.All(),
        ti.Weigh.Equally(),
        ti.Action.Rebalance(),
    ],
    tickers,
)

result = ti.run(ti.Backtest(portfolio, data))

`ti.Not` — invert a signal

ti.Not wraps any single algo and flips its result. The example above shows the most common use: excluding a specific period from an otherwise-regular schedule.

Another example — select only the assets that the momentum filter would not select (i.e. the bottom performers):

ti.Not(
    ti.Select.Momentum(n=3, lookback=pd.DateOffset(months=1), sort_descending=True)
)

Combining all three

Branch operators nest freely. This triggers a rebalance when it is month-end or (it is mid-month and not in a freezing window):

ti.Or(
    ti.Signal.Monthly(),
    ti.And(
        ti.Signal.Monthly(day=15, closest_trading_day=True),
        ti.Not(ti.Signal.Schedule(month=12)),
    ),
)

Note: Sequential algos in the stack are implicitly And — each one must pass for execution to continue. ti.And is only needed when you want to require two conditions at the same stack position.

Fix time rebalance