autoreg.json

[
  {
    "modelParams": {
      "name": "Main",
      "description": "Simple autoregressive model where the expected value of y[n] is α + βy[n−1], with noise scaled as σ",
      "steps": 1,
      "method": "MCMC"
    },
    "blocks": [
      {
        "name": "T",
        "show": true,
        "type": "Data",
        "typeCode": 2,
        "useAsParameter": false,
        "dims": "",
        "value": "1,2.96,7.05,15.1,31.5,64.1,129"
      },
      {
        "distribution": "Uniform",
        "name": "a",
        "once": false,
        "params": {
          "a": "0",
          "b": "20"
        },
        "show": true,
        "type": "Random Variable",
        "typeCode": 0,
        "dims": "1"
      },
      {
        "distribution": "Uniform",
        "name": "b",
        "once": false,
        "params": {
          "a": "0",
          "b": "4"
        },
        "show": true,
        "type": "Random Variable",
        "typeCode": 0,
        "dims": "1"
      },
      {
        "distribution": "Uniform",
        "name": "sigma",
        "once": false,
        "params": {
          "a": "0",
          "b": "5"
        },
        "show": true,
        "type": "Random Variable",
        "typeCode": 0,
        "dims": "1"
      },
      {
        "distribution": "Gaussian",
        "params": {
          "sigma": "sigma",
          "mu": "a + b * T[((_j > 1) ? _j-1 : _j)]"
        },
        "type": "Observer",
        "typeCode": 4,
        "value": "T"
      },
      {
        "distribution": "Gaussian",
        "name": "Tnext",
        "once": false,
        "params": {
          "mu": "a + b * 129",
          "sigma": "sigma"
        },
        "show": true,
        "type": "Random Variable",
        "typeCode": 0,
        "dims": "1"
      }
    ],
    "methodParams": {
      "samples": "2000",
      "burn": "1000",
      "lag": "200"
    }
  }
]