diff --git a/setup.py b/setup.py index a50876e3..6afd3a6d 100644 --- a/setup.py +++ b/setup.py @@ -63,7 +63,7 @@ def run(self): # Run the setup setup( name="tigramite", - version="5.0.1.6", + version="5.0.1.7", packages=["tigramite", "tigramite.independence_tests", "tigramite.toymodels"], license="GNU General Public License v3.0", description="Tigramite causal discovery for time series", diff --git a/tutorials/tigramite_tutorial_general_causal_effect_analysis.ipynb b/tutorials/tigramite_tutorial_general_causal_effect_analysis.ipynb index e62f56c9..da97bbf1 100644 --- a/tutorials/tigramite_tutorial_general_causal_effect_analysis.ipynb +++ b/tutorials/tigramite_tutorial_general_causal_effect_analysis.ipynb @@ -1073,7 +1073,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 20, @@ -1462,7 +1462,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 30, @@ -1491,8 +1491,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "Causal effect for S = -1.00 is 0.02\n", - "Causal effect for S = 1.00 is -0.00\n" + "Causal effect for S = -1.00 is -0.02\n", + "Causal effect for S = 1.00 is 0.00\n" ] } ], @@ -1653,7 +1653,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 34, @@ -1952,7 +1952,7 @@ }, { "cell_type": "code", - "execution_count": 41, + "execution_count": 45, "metadata": {}, "outputs": [ { @@ -1968,11 +1968,16 @@ "\n", "boot_blocklength = 1 \n", "\n", - "(2, 10)\n" + "(2, 1)\n" ] } ], "source": [ + "# Let's generate shorter length data\n", + "T = 100\n", + "data, nonstat = toys.structural_causal_process(links_coeffs, T=T, noises=None, seed=7)\n", + "dataframe = pp.DataFrame(data)\n", + "\n", "# First for the Wright method with effect estimate being 0.75 above\n", "causal_effects.fit_bootstrap_of(\n", " method='fit_wright_effect',\n", @@ -1988,7 +1993,7 @@ }, { "cell_type": "code", - "execution_count": 44, + "execution_count": 51, "metadata": {}, "outputs": [ { @@ -2008,7 +2013,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -2042,7 +2047,8 @@ " method_args={'intervention_data':intervention_data},\n", " conf_lev=0.9)\n", "\n", - "plt.errorbar(intervention_data[:,0], estimated_causal_effects[:,0], estimated_confidence_intervals, label=\"Estimated\")\n", + "plt.errorbar(intervention_data[:,0], estimated_causal_effects[:,0], \n", + " np.abs(estimated_confidence_intervals - estimated_causal_effects[:,0]), label=\"Estimated\")\n", "plt.xlabel('Intervention / X-value range')\n", "plt.ylabel('Causal effect')\n", "plt.legend()\n",