proofreading

demos_ch2/demo2_1.ipynb

@@ -12,7 +12,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "### Bayesian data analysis\n",
+      "### Bayesian Data Analysis, 3rd ed\n",
       "## Chapter 2, demo 1\n",
       "\n",
       "437 girls and 543 boys have been observed. Calculate and plot the posterior distribution of the proportion of girls $\\theta $, using\n",
@@ -115,4 +115,4 @@
    "metadata": {}
   }
  ]
-}
+}

demos_ch2/demo2_1.py

@@ -1,4 +1,4 @@
-"""Bayesian data analysis
+"""Bayesian Data Analysis, 3rd ed
 Chapter 2, demo 1
 
 437 girls and 543 boys have been observed. Calculate and plot the posterior 

demos_ch2/demo2_2.ipynb

@@ -12,10 +12,10 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "### Bayesian data analysis\n",
+      "### Bayesian Data Analysis, 3rd ed\n",
       "## Chapter 2, demo2\n",
       "\n",
-      "Illustrate the effect of prior. Comparison of posterior distributions with different parameter values for the beta prior distribution."
+      "Illustrate the effect of a prior. Comparison of posterior distributions with different parameter values for Beta prior distribution."
      ]
     },
     {
@@ -143,4 +143,4 @@
    "metadata": {}
   }
  ]
-}
+}

demos_ch2/demo2_2.py

@@ -1,8 +1,8 @@
-"""Bayesian data analysis
+"""Bayesian data analysis, 3rd ed
 Chapter 2, demo 2
 
-Illustrate the effect of prior. Comparison of posterior distributions with 
-different parameter values for the beta prior distribution.
+Illustrate the effect of a prior. Comparison of posterior distributions with 
+different parameter values for Beta prior distribution.
 
 """
 

demos_ch2/demo2_3.ipynb

@@ -12,7 +12,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "### Bayesian data analysis\n",
+      "### Bayesian Data Analysis, 3rd ed\n",
       "## Chapter 2, demo3\n",
       "\n",
       "Simulate samples from Beta(438,544), draw a histogram with quantiles, and do the same for a transformed variable."
@@ -112,4 +112,4 @@
    "metadata": {}
   }
  ]
-}
+}

demos_ch2/demo2_3.py

@@ -1,4 +1,4 @@
-"""Bayesian data analysis
+"""Bayesian Data Analysis, 3rd ed
 Chapter 2, demo 3
 
 Simulate samples from Beta(438,544), draw a histogram with quantiles, and do 

demos_ch2/demo2_4.ipynb

@@ -12,7 +12,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "### Bayesian data analysis\n",
+      "### Bayesian Data Analysis, 3rd ed\n",
       "## Chapter 2, demo4\n",
       "\n",
       "Calculate the posterior distribution on a discrete grid of points by multiplying the likelihood and a non-conjugate prior at each point, and normalizing over the points. Simulate samples from the resulting non-standard posterior distribution using inverse cdf using the discrete grid."
@@ -72,7 +72,7 @@
       "x = np.linspace(0, 1, nx)\n",
       "\n",
       "# Compute density of non-conjugate prior in grid\n",
-      "# This non-conjugate prior is same as in figure 2.4 in the book\n",
+      "# This non-conjugate prior is same as in Figure 2.4 in the book\n",
       "pp = np.ones(nx)\n",
       "ascent = (0.385 <= x) & (x <= 0.485)\n",
       "descent = (0.485 <= x) & (x <= 0.585)\n",
@@ -82,7 +82,7 @@
       "# Normalize the prior\n",
       "pp /= np.sum(pp)\n",
       "\n",
-      "# Unnormalsed non-conjugate posterior in grid\n",
+      "# Unnormalised non-conjugate posterior in grid\n",
       "po = beta.pdf(x, a, b)*pp\n",
       "po /= np.sum(po)\n",
       "# Cumulative\n",
@@ -184,4 +184,4 @@
    "metadata": {}
   }
  ]
-}
+}

demos_ch2/demo2_4.py

@@ -1,4 +1,4 @@
-"""Bayesian data analysis
+"""Bayesian Data Analysis, 3rd ed
 Chapter 2, demo 4
 
 Calculate the posterior distribution on a discrete grid of points by 
@@ -28,7 +28,7 @@
 x = np.linspace(0, 1, nx)
 
 # Compute density of non-conjugate prior in grid
-# This non-conjugate prior is same as in figure 2.4 in the book
+# This non-conjugate prior is same as in Figure 2.4 in the book
 pp = np.ones(nx)
 ascent = (0.385 <= x) & (x <= 0.485)
 descent = (0.485 <= x) & (x <= 0.585)
@@ -38,7 +38,7 @@
 # Normalize the prior
 pp /= np.sum(pp)
 
-# Unnormalsed non-conjugate posterior in grid
+# Unnormalised non-conjugate posterior in grid
 po = beta.pdf(x, a, b)*pp
 po /= np.sum(po)
 # Cumulative

demos_ch3/demo3_1-4.ipynb

@@ -11,11 +11,11 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "### Bayesian data analysis\n",
-      "##  Chapter 3, demo 1-4\n",
+      "### Bayesian Data Analysis, 3rd ed\n",
+      "##  Chapter 3, demos 1-4\n",
       "\n",
-      "Examples and illustrations for normal model with unknown mean and variance\n",
-      "(BDA section 3.2 on p. 64)."
+      "Examples and illustrations for a normal model with unknown mean and variance\n",
+      "(BDA3 section 3.2 on p. 64)."
      ]
     },
     {
@@ -116,7 +116,7 @@
       "tlynew = [50, 185]\n",
       "xynew = np.linspace(tlynew[0], tlynew[1], 1000)\n",
       "\n",
-      "# evaluate joint density in grid\n",
+      "# evaluate the joint density in a grid\n",
       "# note that the following is not normalized, but for plotting\n",
       "# contours it does not matter\n",
       "Z = stats.norm.pdf(t1, my, t2[:,np.newaxis]/np.sqrt(n))\n",
@@ -127,7 +127,7 @@
       "# z=(x-mean(y))/sqrt(s2/n), see BDA3 p. 21\n",
       "pm_mu = stats.t.pdf((t1 - my) / np.sqrt(s2/n), n-1) / np.sqrt(s2/n)\n",
       "\n",
-      "# estimate the marginal density for mu using samples and ad hoc Gaussian\n",
+      "# estimate the marginal density for mu using samples and an ad hoc Gaussian\n",
       "# kernel approximation\n",
       "pk_mu = stats.gaussian_kde(mu).evaluate(t1)\n",
       "\n",
@@ -137,7 +137,7 @@
       "pm_sigma = sinvchi2.pdf(t2**2, n-1, s2)*2*t2\n",
       "# N.B. this was already calculated in the joint distribution case\n",
       "\n",
-      "# estimate the marginal density for sigma using samples and ad hoc Gaussian\n",
+      "# estimate the marginal density for sigma using samples and an ad hoc Gaussian\n",
       "# kernel approximation\n",
       "pk_sigma = stats.gaussian_kde(sigma).evaluate(t2)\n",
       "\n",
@@ -155,7 +155,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "Visualise joint density and marginal densities of posterior of normal \n",
+      "Visualise the joint density and marginal densities of the posterior of normal \n",
       "distribution with unknown mean and variance."
      ]
     },
@@ -167,13 +167,13 @@
       "plotgrid = gridspec.GridSpec(2, 2, width_ratios=[3,2], height_ratios=[3,2])\n",
       "fig = plt.figure(figsize=(10,10))\n",
       "\n",
-      "# plot joint distribution\n",
+      "# plot the joint distribution\n",
       "plt.subplot(plotgrid[0,0])\n",
       "# plot the contour plot of the exact posterior (c_levels is used to give\n",
       "# a vector of linearly spaced values at which levels contours are drawn)\n",
       "c_levels = np.linspace(1e-5, Z.max(), 6)[:-1]\n",
       "plt.contour(t1, t2, Z, c_levels, colors='blue')\n",
-      "# plot samples from the joint posterior\n",
+      "# plot the samples from the joint posterior\n",
       "samps = plt.scatter(mu, sigma, 5, color=[0.25, 0.75, 0.25])\n",
       "# decorate\n",
       "plt.xlim(tl1)\n",
@@ -187,7 +187,7 @@
       "    loc='upper center', \n",
       ")\n",
       "\n",
-      "# plot marginal of mu\n",
+      "# plot the marginal of mu\n",
       "plt.subplot(plotgrid[1,0])\n",
       "# empirical\n",
       "plt.plot(t1, pk_mu, color='#ff8f20', linewidth=2.5, label='empirical')\n",
@@ -199,7 +199,7 @@
       "plt.yticks(())\n",
       "plt.legend()\n",
       "\n",
-      "# plot marginal of sigma\n",
+      "# plot the marginal of sigma\n",
       "plt.subplot(plotgrid[0,1])\n",
       "# empirical\n",
       "plt.plot(pk_sigma, t2, color='#ff8f20', linewidth=2.5, label='empirical')\n",
@@ -230,7 +230,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "Visualise factored sampling and corresponding marginal and conditional density."
+      "Visualise factored sampling and the corresponding marginal and conditional densities."
      ]
     },
     {
@@ -241,7 +241,7 @@
       "plotgrid = gridspec.GridSpec(1, 2, width_ratios=[3,2])\n",
       "fig = plt.figure(figsize=(10,7))\n",
       "\n",
-      "# plot joint distribution\n",
+      "# plot the joint distribution\n",
       "ax0 = plt.subplot(plotgrid[0,0])\n",
       "# plot the contour plot of the exact posterior (c_levels is used to give\n",
       "# a vector of linearly spaced values at which levels contours are drawn)\n",
@@ -272,7 +272,7 @@
       "    fontsize=14\n",
       ")\n",
       "\n",
-      "# plot marginal of sigma\n",
+      "# plot the marginal of sigma\n",
       "ax1 = plt.subplot(plotgrid[0,1])\n",
       "plt.plot(pm_sigma, t2, 'b', linewidth=1.5)\n",
       "# decorate\n",
@@ -298,7 +298,7 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "Visualise marginal distribution of mu as a mixture of normals."
+      "Visualise the marginal distribution of mu as a mixture of normals."
      ]
     },
     {
@@ -360,13 +360,13 @@
       "\n",
       "fig = plt.figure(figsize=(11,11))\n",
       "\n",
-      "# plot joint distribution\n",
+      "# plot the joint distribution\n",
       "plt.subplot(2,2,1)\n",
       "# plot the contour plot of the exact posterior (c_levels is used to give\n",
       "# a vector of linearly spaced values at which levels contours are drawn)\n",
       "c_levels = np.linspace(1e-5, Z.max(), 6)[:-1]\n",
       "plt.contour(t1, t2, Z, c_levels, colors='blue')\n",
-      "# plot samples from the joint posterior\n",
+      "# plot the samples from the joint posterior\n",
       "samps = plt.scatter(mu, sigma, 5, color=[0.25, 0.75, 0.25])\n",
       "# decorate\n",
       "plt.xlim(tl1)\n",
@@ -385,7 +385,7 @@
       "\n",
       "# plot first ynew\n",
       "plt.subplot(2,2,2)\n",
-      "# plot first distribution and the respective sample\n",
+      "# plot the distribution and the respective sample\n",
       "line1, = plt.plot(xynew, ynewdists[0], 'b', linewidth=1.5)\n",
       "ax1_hs = plt.scatter(ynew[0], 0.02*np.max(ynewdists), 40, 'r')\n",
       "# decorate\n",
@@ -434,4 +434,4 @@
    "metadata": {}
   }
  ]
-}
+}

demos_ch3/demo3_1.py

@@ -1,7 +1,7 @@
-"""Bayesian data analysis
+"""Bayesian Data Analysis, 3r ed
 Chapter 3, demo 1
 
-Visualise joint density and marginal densities of posterior of normal 
+Visualise the joint density and marginal densities of posterior of normal 
 distribution with unknown mean and variance.
 
 """
@@ -51,7 +51,7 @@
 tl2 = [10, 60]
 t2 = np.linspace(tl2[0], tl2[1], 1000)
 
-# evaluate joint density in grid
+# evaluate the joint density in a grid
 # note that the following is not normalized, but for plotting
 # contours it does not matter
 Z = stats.norm.pdf(t1, my, t2[:,np.newaxis]/np.sqrt(n))
@@ -62,8 +62,8 @@
 # z=(x-mean(y))/sqrt(s2/n), see BDA3 p. 21
 pm_mu = stats.t.pdf((t1 - my) / np.sqrt(s2/n), n-1) / np.sqrt(s2/n)
 
-# estimate the marginal density for mu using samples and ad hoc Gaussian
-# kernel approximation
+# estimate the marginal density for mu using samples and an ad hoc
+# Gaussian kernel approximation
 pk_mu = stats.gaussian_kde(mu).evaluate(t1)
 
 # compute the exact marginal density for sigma
@@ -72,7 +72,7 @@
 pm_sigma = sinvchi2.pdf(t2**2, n-1, s2)*2*t2
 # N.B. this was already calculated in the joint distribution case
 
-# estimate the marginal density for sigma using samples and ad hoc Gaussian
+# estimate the marginal density for sigma using samples and an ad hoc Gaussian
 # kernel approximation
 pk_sigma = stats.gaussian_kde(sigma).evaluate(t2)
 
@@ -83,13 +83,13 @@
 plotgrid = gridspec.GridSpec(2, 2, width_ratios=[3,2], height_ratios=[3,2])
 plt.figure(figsize=(12,12))
 
-# plot joint distribution
+# plot the joint distribution
 plt.subplot(plotgrid[0,0])
 # plot the contour plot of the exact posterior (c_levels is used to give
 # a vector of linearly spaced values at which levels contours are drawn)
 c_levels = np.linspace(1e-5, Z.max(), 6)[:-1]
 plt.contour(t1, t2, Z, c_levels, colors='blue')
-# plot samples from the joint posterior
+# plot the samples from the joint posterior
 samps = plt.scatter(mu, sigma, 5, color=[0.25, 0.75, 0.25])
 # decorate
 plt.xlim(tl1)
@@ -102,7 +102,7 @@
     ('exact contour plot', 'samples')
 )
 
-# plot marginal of mu
+# plot the marginal of mu
 plt.subplot(plotgrid[1,0])
 # empirical
 plt.plot(t1, pk_mu, color='#ff8f20', linewidth=2.5, label='empirical')
@@ -114,7 +114,7 @@
 plt.yticks(())
 plt.legend()
 
-# plot marginal of sigma
+# plot the marginal of sigma
 plt.subplot(plotgrid[0,1])
 # empirical
 plt.plot(pk_sigma, t2, color='#ff8f20', linewidth=2.5, label='empirical')