Fix arch_model vol check and expand FIAPARCH reduced-order tests

Fix bug using vol_model instead of vol in arch_model type check.
Add tests for FIAPARCH with reduced p, o, q combinations including
forecasting, simulation, parameter names, and backcast_transform.

Made-with: Cursor

Author: gjunjie

arch/tests/univariate/test_mean.py

@@ -1363,6 +1363,18 @@ def test_fiaparch_non_int_p():
         arch_model(SP500, vol="fiaparch", p=[1, 2])
 
 
+def test_fiaparch_arch_model_kwargs():
+    am = arch_model(SP500, vol="fiaparch", o=0)
+    assert isinstance(am.volatility, FIAPARCH)
+    assert am.volatility.o == 0
+    assert am.volatility.name == "FI Power ARCH"
+
+    am2 = arch_model(SP500, vol="FIAPARCH", p=0, q=0)
+    assert isinstance(am2.volatility, FIAPARCH)
+    assert am2.volatility.p == 0
+    assert am2.volatility.q == 0
+
+
 def test_param_cov():
     mod = ConstantMean(SP500)
     res = mod.fit(disp="off")

arch/tests/univariate/test_variance_forecasting.py

@@ -2010,6 +2010,62 @@ def test_fiaparch_simulation_smoke(self, o, delta):
                 method="analytic",
             )
 
+    @pytest.mark.parametrize("p,q", [(0, 1), (1, 0), (0, 0)])
+    def test_fiaparch_one_step_reduced(self, p, q):
+        trunc = 50
+        vol = FIAPARCH(p=p, q=q, truncation=trunc)
+        resids = self.resid
+        backcast = vol.backcast(resids)
+        var_bounds = vol.variance_bounds(resids)
+        params = [0.1]
+        if p:
+            params.append(0.2)
+        params.append(0.4)
+        if q:
+            params.append(0.2)
+        params.extend([-0.3, 1.5])
+        params = np.array(params)
+        sigma2 = np.empty_like(resids)
+        vol.compute_variance(params, resids, sigma2, backcast, var_bounds)
+        forecast = vol.forecast(
+            params, resids, backcast, var_bounds, horizon=1, start=0
+        )
+        assert_allclose(sigma2[1:], forecast.forecasts[:-1, 0])
+
+    @pytest.mark.parametrize("p,q", [(0, 1), (1, 0), (0, 0)])
+    def test_fiaparch_simulation_reduced(self, p, q):
+        dist = Normal(seed=self.rng)
+        rng = dist.simulate([])
+        trunc = 50
+        vol = FIAPARCH(p=p, q=q, truncation=trunc)
+        resids = self.resid
+        backcast = vol.backcast(resids)
+        var_bounds = vol.variance_bounds(resids)
+        params = [0.1]
+        if p:
+            params.append(0.2)
+        params.append(0.4)
+        if q:
+            params.append(0.2)
+        params.extend([-0.3, 1.5])
+        params = np.array(params)
+        sigma2 = np.empty_like(resids)
+        vol.compute_variance(params, resids, sigma2, backcast, var_bounds)
+        forecast = vol.forecast(
+            params,
+            resids,
+            backcast,
+            var_bounds,
+            horizon=10,
+            start=0,
+            method="simulation",
+            rng=rng,
+            simulations=100,
+        )
+        assert_allclose(sigma2[1:], forecast.forecasts[:-1, 0])
+        assert forecast.forecast_paths is not None
+        assert forecast.shocks is not None
+
     def test_midas_analytical(self):
         vol = MIDASHyperbolic()
         resids = self.resid

arch/tests/univariate/test_volatility.py

@@ -1979,11 +1979,23 @@ def test_fiaparch_no_phi(setup):
     # omega, d, beta, gamma, delta
     assert len(bounds) == fiaparch.num_params
 
+    names = fiaparch.parameter_names()
+    assert "phi" not in names
+    assert names == ["omega", "d", "beta", "gamma", "delta"]
+
     a, _ = fiaparch.constraints()
     # No phi => FIGARCH block shrinks (5 rows), gamma (2 rows), delta (2 rows) = 9
     assert a.shape[1] == 5
     assert a.shape[0] == 9
 
+    backcast = fiaparch.backcast(setup.resids)
+    var_bounds = fiaparch.variance_bounds(setup.resids)
+    # omega, d, beta, gamma, delta (no phi)
+    parameters = np.array([1.0, 0.4, 0.2, -0.3, 1.5])
+    sigma2 = np.zeros_like(setup.sigma2)
+    fiaparch.compute_variance(parameters, setup.resids, sigma2, backcast, var_bounds)
+    assert np.all(np.isfinite(sigma2))
+
 
 def test_fiaparch_no_beta(setup):
     fiaparch = FIAPARCH(q=0)
@@ -1995,6 +2007,10 @@ def test_fiaparch_no_beta(setup):
     # omega, phi, d, gamma, delta
     assert len(bounds) == fiaparch.num_params
 
+    names = fiaparch.parameter_names()
+    assert "beta" not in names
+    assert names == ["omega", "phi", "d", "gamma", "delta"]
+
     a, _ = fiaparch.constraints()
     # No beta => FIGARCH block shrinks (5 rows), gamma (2 rows), delta (2 rows) = 9
     assert a.shape[1] == 5
@@ -2022,6 +2038,7 @@ def test_fiaparch_no_asym(setup):
 
     names = fiaparch.parameter_names()
     assert "gamma" not in names
+    assert names == ["omega", "phi", "d", "beta", "delta"]
 
     assert_equal(fiaparch.name, "FI Power ARCH")
 
@@ -2124,6 +2141,30 @@ def test_fiaparch_simulate_high_beta(setup):
             fiaparch.simulate(parameters, 20, rng.simulate([]))
 
 
+@pytest.mark.parametrize(
+    "p,o,q",
+    [(0, 1, 1), (1, 0, 1), (1, 1, 0), (0, 0, 1), (0, 1, 0), (1, 0, 0), (0, 0, 0)],
+)
+def test_fiaparch_simulate_reduced(setup, p, o, q):
+    fiaparch = FIAPARCH(p=p, o=o, q=q, truncation=100)
+    rng = Normal(seed=RandomState())
+    params = [1.0]
+    if p:
+        params.append(0.2)
+    params.append(0.4)
+    if q:
+        params.append(0.2)
+    if o:
+        params.append(-0.3)
+    params.append(1.5)
+    parameters = np.array(params)
+    sim_data = fiaparch.simulate(parameters, setup.t, rng.simulate([]))
+    assert sim_data[0].shape[0] == setup.t
+    assert sim_data[1].shape[0] == setup.t
+    assert np.all(np.isfinite(sim_data[0]))
+    assert np.all(np.isfinite(sim_data[1]))
+
+
 def test_fiaparch_backcast_transform(setup):
     fiaparch = FIAPARCH()
     backcast = fiaparch.backcast(setup.resids)
@@ -2137,6 +2178,48 @@ def test_fiaparch_backcast_transform(setup):
     assert result_arr.shape == (2,)
     assert_allclose(result_arr, np.array([expected, expected]))
 
+    fiaparch_fd = FIAPARCH(delta=1.5)
+    backcast_fd = fiaparch_fd.backcast(setup.resids)
+    result_fd = fiaparch_fd.backcast_transform(backcast_fd)
+    assert np.isscalar(result_fd)
+    expected_fd = float(np.sqrt(backcast_fd) ** 1.5)
+    assert_allclose(result_fd, expected_fd)
+
+    result_fd_arr = fiaparch_fd.backcast_transform(
+        np.array([backcast_fd, backcast_fd])
+    )
+    assert result_fd_arr.shape == (2,)
+    assert_allclose(result_fd_arr, np.array([expected_fd, expected_fd]))
+
+
+def test_fiaparch_minimal(setup):
+    fiaparch = FIAPARCH(p=0, q=0)
+
+    assert fiaparch.num_params == 4  # omega, d, gamma, delta
+
+    names = fiaparch.parameter_names()
+    assert names == ["omega", "d", "gamma", "delta"]
+
+    bounds = fiaparch.bounds(setup.resids)
+    assert len(bounds) == 4
+
+    a, b = fiaparch.constraints()
+    # FIGARCH block: omega > 0, d > 0, d < 1 => 3 rows
+    # gamma: 2 rows, delta: 2 rows => total 7
+    assert a.shape == (7, 4)
+    assert b.shape == (7,)
+
+    sv = fiaparch.starting_values(setup.resids)
+    assert sv.shape[0] == 4
+
+    backcast = fiaparch.backcast(setup.resids)
+    var_bounds = fiaparch.variance_bounds(setup.resids)
+    # omega, d, gamma, delta
+    parameters = np.array([1.0, 0.4, -0.3, 1.5])
+    sigma2 = np.zeros_like(setup.sigma2)
+    fiaparch.compute_variance(parameters, setup.resids, sigma2, backcast, var_bounds)
+    assert np.all(np.isfinite(sigma2))
+
 
 def test_fiaparch_errors(setup):
     with pytest.raises(ValueError, match=r"truncation must be a positive integer"):
@@ -2165,3 +2248,4 @@ def test_fiaparch_str(setup, p, o, q):
     assert f"q: {q}" in s
     assert f"p: {p}" in s
     assert f"o: {o}" in s
+    assert "delta:" not in s

arch/univariate/mean.py

@@ -2040,7 +2040,7 @@ def arch_model(
     else:  # mean == "zero"
         am = ZeroMean(y, hold_back=hold_back, rescale=rescale)
 
-    if vol_model in ("arch", "garch", "figarch", "fiaparch", "egarch", "aparch") and not isinstance(
+    if vol in ("arch", "garch", "figarch", "fiaparch", "egarch", "aparch") and not isinstance(
         p, int
     ):
         raise TypeError(