# 引用 - 高级函数

## AND

NOTE
0（零）表示 False，而任何其他值均表示 True。
``````AND(logical_test1,[logical_test2],...)
``````

logical_test1

logical_test2

## 非重复近似计数（维度）

``````Approximate Count Distinct (dimension)
``````

dimension

## 用例示例

Approximate Count Distinct (customer ID eVar) 是此函数的常见用例。

## 比较计数函数

Approximate Count Distinct() 是对 Count() 和 RowCount() 函数所做出的改进，因为创建的量度可用于任何维度报表，以呈现单独维度项目的近似计数。例如，“移动设备类型”报表中使用的客户 ID 计数。

## 反余弦 (Row)

``````ACOS(metric)
``````

metric

## 反正弦 (Row)

``````ASIN(metric)
``````

metric

## 反正切 (Row)

``````ATAN(metric)
``````

metric

## 指数回归：预测的 Y (Row)

``````ESTIMATE.EXP(metric_X, metric_Y)
``````

metric_X

metric_Y

## Cdf-T

``````cdf_t( -∞, n ) = 0
cdf_t(  ∞, n ) = 1
cdf_t( 3, 5 ) ? 0.99865
cdf_t( -2, 7 ) ? 0.0227501
cdf_t( x, ∞ ) ? cdf_z( x )
``````

## Cdf-Z

``````cdf_z( -∞ ) = 0
cdf_z( ∞ ) = 1
cdf_z( 0 ) = 0.5
cdf_z( 2 ) ? 0.97725
cdf_z( -3 ) ? 0.0013499
``````

## 向上取整 (Row)

``````CEILING(metric)
``````

metric

## 置信度

``````fx Confidence (normalizing-container, success-metric, control, significance-threshold)
``````

## 余弦 (Row)

``````COS(metric)
``````

metric

## 立方根

``````CBRT(metric)
``````

metric

## 累积

``````| Date | Rev  | cumul(0,Rev) | cumul(2,Rev) |
|------+------+--------------+--------------|
| May  | \$500 | \$500         | \$500         |
| June | \$200 | \$700         | \$700         |
| July | \$400 | \$1100        | \$600         |
``````

## 累积平均数

NOTE

``````cumul(revenue)/cumul(person)
``````

## 指数回归_ 相关系数 (Table)

``````CORREL.EXP(metric_X, metric_Y)
``````

metric_X

metric_Y

## 指数回归：截距 (Table)

``````INTERCEPT.EXP(metric_X, metric_Y)
``````

metric_X

metric_Y

## 指数回归：斜率 (Table)

``````SLOPE.EXP(metric_X, metric_Y)
``````

metric_X

metric_Y

## 向下取整 (Row)

``````FLOOR(metric)
``````

metric

## 双曲余弦 (Row)

``````COSH(metric)
``````

metric

## 双曲正弦 (Row)

``````SINH(metric)
``````

metric

## 双曲正切 (Row)

``````TANH(metric)
``````

metric

## IF (Row)

IF 函数可在您指定的条件计算为 TRUE 时返回一个值，在该条件计算为 FALSE 时返回另一个值。

``````IF(logical_test, [value_if_true], [value_if_false])
``````

logical_test

[value_if_true]

[value_if_false]

## 提升

``````fx Lift (normalizing-container, success-metric, control)
``````

## 线性回归_ 相关系数

Y = a X + b。返回相关系数。

## 线性回归_ 截距

Y = a X + b。返回 b。

## 线性回归_ 预测的 Y

Y = a X + b。返回 Y。

## 线性回归_ 斜率

Y = a X + b。返回 a。

## 以 10 为底的对数 (Row)

``````LOG10(metric)
``````

metric

## 对数回归：相关系数 (Table)

``````CORREL.LOG(metric_X,metric_Y)
``````

metric_X

metric_Y

## 对数回归：截距 (Table)

``````INTERCEPT.LOG(metric_X, metric_Y)
``````

metric_X

metric_Y

## 对数回归：预测的 Y（行）

``````ESTIMATE.LOG(metric_X, metric_Y)
``````

metric_X

metric_Y

## 对数回归：斜率 (Table)

``````SLOPE.LOG(metric_A, metric_B)
``````

metric_A

metric_B

## 自然对数

``````LN(metric)
``````

metric

## NOT

``````NOT(logical)
``````

logical

## 或 (Row)

NOTE
0（零）表示 False，而任何其他值均表示 True。
``````OR(logical_test1,[logical_test2],...)
``````

logical_test1

logical_test2

## Pi

``````PI()
``````

PI 函数没有参数。

## 幂回归：相关系数 (Table)

``````CORREL.POWER(metric_X, metric_Y)
``````

metric_X

metric_Y

## 幂回归：截距 (Table)

`````` INTERCEPT.POWER(metric_X, metric_Y)
``````

metric_X

metric_Y

## 幂回归：预测的 Y (Row)

`````` ESTIMATE.POWER(metric_X, metric_Y)
``````

metric_X

metric_Y

## 幂回归：斜率 (Table)

``````SLOPE.POWER(metric_X, metric_Y)
``````

metric_X

metric_Y

## 二次回归：相关系数 (Table)

``````CORREL.QUADRATIC(metric_X, metric_Y)
``````

metric_X

metric_Y

## 二次回归：截距 (Table)

``````INTERCEPT.POWER(metric_X, metric_Y)
``````

metric_X

metric_Y

## 二次回归：预测的 Y (Row)

``````ESTIMATE.QUADRATIC(metric_A, metric_B)
``````

metric_A

metric_B

## 二次回归：斜率 (Table)

``````SLOPE.QUADRATIC(metric_X, metric_Y)
``````

metric_X

metric_Y

## 倒数回归：相关系数 (Table)

``````CORREL.RECIPROCAL(metric_X, metric_Y)
``````

metric_X

metric_Y

## 倒数回归：截距 (Table)

``````INTERCEPT.RECIPROCAL(metric_A, metric_B)
``````

metric_X

metric_Y

## 倒数回归：预测的 Y (Row)

``````ESTIMATE.RECIPROCAL(metric_X, metric_Y)
``````

metric_X

metric_Y

## 倒数回归：斜率 (Table)

``````SLOPE.RECIPROCAL(metric_X, metric_Y)
``````

metric_X

metric_Y

## 正弦 (Row)

``````SIN(metric)
``````

metric

## T 分数

Z 分数的别名，即由平均值偏差除以标准偏差

## T 测试

`X` 为 t 测试统计数据，通常是基于量度的公式（例如 zscore），每行都将对其进行评估。

1. 用其查找离群值：

code language-none
``````t_test( zscore(bouncerate), row-count-1, 2)
``````
2. 将其与 `if` 结合使用，以忽略过高或过低的跳出率，并对其他内容的访问次数进行计数：

code language-none
``````if ( t_test( z-score(bouncerate), row-count, 2) < 0.01, 0, visits )
``````

## 正切

``````TAN (metric)
``````

metric

## Z 分数 (Row)

Z 分数的方程式为：

NOTE
μ (mu) 和 σ (sigma) 会使用该量度自动计算。

Z 分数（量度）

metric

## Z 测试

NOTE

recommendation-more-help
080e5213-7aa2-40d6-9dba-18945e892f79