Since the training and testing of neural networks is **very data-intensive**, this poses serious obstacles to the development of neural network applications in ecology.

Michaelsen J. 1987. Cross-validation in statistical climate forecast models.

`1520-0450(1987)026-1589-cviscf-2.0.co;2.pdf`

Set consists of predictions and targets

A set of prediction rule will be used to predict y_{0} from

Let be the accuracy.

by least squares this will usually

in other words expected Err is

In cross validation

statistical error is then

while residual is

hat over the letter ε indicates an observable estimate of an **unobservable quantity** called ε.

no |
datum |
drawn as |

1 | 0.026 | outlier |

2 | 0.048 | whisker low |

3 | 0.070 | Q1 |

4 | 0.072 | |

5 | 0.076 | Q2 |

6 | 0.084 | |

7 | 0.086 | |

8 | 0.099 | Q3 |

9 | 0.102 | |

10 | 0.103 | whisker high |

0.026 becomes **outlier** because 1.5IQR boundary (IQR = Q3 – Q1 = 0.02908) for the lower boundary is Q1 -1.5IQR = 0.0404, hence whisker is 0.048 which is more close to the box inside making 0.026 outlier.

The cgs physical unit for dynamic viscosity is the poise (P), more commonly expressed, particularly in ASTM standards, as centipoise (cP)

]]>