2008年5月31日星期六

Mark 2

Chapter 3 is about conflicts and Jesus' reaction

Jesus' enemy want to entrap Jesus, bringing a man with a shriveled hand before him on the Sabbath. To abide the tradition, he cannot heal the man. Jesus tried to provoke his enemy's conscience, yet it does not work. He healed. His enemy got the excuse to kill him. He withdrew to the lake and appointed twelve apostles to defend his church.

His family disliked him and tried to take charge of him. He denied his mother and brothers because they did not follow God.

Here is one difficulty of Christianity in China. Chinese tradition gave parents authority. It is hard for Chinese to love God more than their parents.

Even he was not in Jerusalem, his enemy condemned him as an evil spirit. Jesus defended himself using several parables. Evil cannot drive out evil, and only holy spirit can tie up evil. In his eyes, evils must be the same even there are different kinds of evils.

Chapter 4 Parables

Sowing seeds and different results.

"For whatever is hidden is meant to be disclosed, and whatever is concealed is meant to be brought out into the open."

Jesus calms the storm.

2008年5月30日星期五

Mark 1

"Mark" is a brief story of Jesus Christ. Because it is brief, for me, a beginner, it is a good primary textbook.

Chapter 1 is about that Jesus is omnipotent. He healed many and drove out evil spirits. He has great influence, because it seems that he can call disciples before he showed his power.

But there are two questions:

Jesus was baptized by another preacher, John. Why there is a John before Jesus? Is John a teacher of Jesus? It seems that he is not one of Jesus' disciples. Is he involved in Jesus' activities following?

Answer: John is said to be the one who prepared the way for Jesus. He is not a teacher of Jesus. He is not a disciple of Jesus. Yes, he is in the following activities, see Luke.

In 1:35-39, Jesus left his disciples and prayed in a solitary place early in the morning? Why he did that? When his disciples found him, why he decided to go somewhere else?

Answer: Early morning and praying alone is good to plan for the whole day. After thinking, Jesus decided to build up ministries.

In chapter 2 Jesus was challenged four times.

The first one is when Jesus healed a paralytic. At first he said "your sins are forgiven." Some teachers of law thought that Jesus is blaspheming in their mind. Jesus answered them by healing this paralytic to show that he is the lord and there is no blasphemy. It is a quiet fight between Jesus and his enemies.

The second one is when Jesus was invited to dinner with tax collectors and sinners. Someone asked his disciples why Jesus is banded with these guys. Jesus told them he is a doctor and he comes to sinners but not the righteous. This is an indirect fight.

The third time is about fasting. Jesus' disciples did not fast. Someone asked Jesus why. Jesus answered by two metaphors. He is the lord, the lord need not fast, and so do his disciples when they are together with the lord. He is new, different from other preachers. So it is never a good idea to attach old things to something new. This is a direct fight, but Jesus used metaphors.

The forth one is about his disciples did something forbidden on the Sabbath. Someone questioned Jesus why they did unlawful things. Jesus gave them an example that David did something forbidden before. Because of his dignity, it is not a problem at all. This is a totally direct fight, no metaphors.

These four stories tell me that Jesus is audacious and righteous. Because he and his disciples did many things which are strange for common people, he is audacious. Because he always answered question, regardless of direct or indirect questions, directly to the people who questioned him, he is righteous.

2008年5月28日星期三

Go to hell, Lang Xianping.

I read an essay by Lang Xianping last night. This morning it is forwarded by a friend of mine. Reading something twice can encourage the reader to write a comment,and here it is.

I know little about Lang Xianping. Yet I still feel him disgusting. It
seems that he is preaching that Chinese need believe in God, or
anything like that. Otherwise Chinese does not deserve modern
enterprises. Basically Chinese cannot benefit from modern enterprises
because of the culture. If Chinese can benefit, they must be
hypocrisy. And the reason why American or European people are more
wealthy is that they are more moral than Chinese. I do not mean that
Chinese do not have any problem in morality. But I think blame the
dysfunction for firm owners are all immoral is not the point. Lang
must be more a propaganda machine than a serious and righteous
economist. He should not be blind at all kinds of shortfalls of
current system, but blame for education or morality. If he is right,
then Adam Smith's invisible hands must be wrong.

I can imagine that in his ideal world, it must be a Communism, and Lei
Feng could be an ideal entrepreneur. Probably he think Cultural
Revolution in China is even righteous.

2008年5月8日星期四

我常用到的stata命令之六

Stata的估计和检验功能强大,尤其是处理微观数据。回归是实证研究的常用方法。而线性设定下的最小二乘法(OLS)和两阶段最小二乘法(2SLS)又是应用最广泛的,应该优先掌握。先讲如何在stata中实现OLS和2SLS估计,再讲如何在选择合理的方法处理实际问题。后一部分深受Joshua Angrist教授影响,在此致谢,后面引用他的思想时会一一注明。

估计的命令简单明了,就不多说了。

reg (被解释变量) (解释变量1) (解释变量2)……

执行上面的命令后,stata会报告回归的估计结果和一些方差分析。在下方的表格中依次列出了解释变量,其系数估计值,估计值的标准误,t比率,p值,及置信度为(1-0.05)的置信区间。

拿到回归结果后,先看看符号对不对,是否显著。解释变量影响的方向和大小直接从点估计值读出,显著性由t统计量得出。在经典假设下,t比率服从t分布。t分布和标准正态分布形状相似,只是它的“尾巴”要比标准正态分布的“肥”一些。当样本量趋于无穷时,t分布的极限分布是标准正态分布,因而“肥尾”的特征逐渐消失。下表列出了不同自由度下二者的差异(Beyer 1987 “CRC Standard Mathematical Tables, 28th ed.”;Goulden 1956 “Methods of Statistical Analysis, 2nd ed.”)。可以看出,自由度超过一百时,二者的差别就已经相当小了。所以,当样本量的数量级是100或以上时,用标准正态分布的关键点作检验是比较准确的。比如,1.96是97.5%的关键点,1.64是95%的关键点。这些都是比较熟悉的。


90% 95% 97.5% 99.5%
1 3.07768 6.31375 12.7062 63.6567
2 1.88562 2.91999 4.30265 9.92484
3 1.63774 2.35336 3.18245 5.84091
4 1.53321 2.13185 2.77645 4.60409
5 1.47588 2.01505 2.57058 4.03214
10 1.37218 1.81246 2.22814 3.16927
30 1.31042 1.69726 2.04227 2.75000
100 1.29007 1.66023 1.98397 2.62589

1.28156 1.64487 1.95999 2.57588

读者读到这里可能会笑话我了。stata不是已经报告了t检验的p值和置信区间了吗?为什么不直接察看这些结果呢?原因在于实证文献往往只报告参数的估计值和标准误,需要读者自己将估计值和标准误相除来计算显著性。而且当你在写实证文章时,也应该报告参数的估计值和标准误。因为p值和置信区间是基于待估计参数等于零的原假设的,如果读者要做其他原假设并不是系数等于零的检验,或者单尾检验时,知道标准误就很方便。所以,报告标准误比报告p值更好。

和回归命令相伴使用的一个重要命令是predict。它的用处是在回归结束后得到相关的统计量。语法如下:

predict (新变量名), (统计量名)

这里的统计量名是一些选项。常用的选项有:xb(回归的拟合值。这是默认选项,即不加任何选项时,predict会给新变量赋一个拟合值。);residuals(残差);leverage(杠杆值)。后面举一个例子。
(待续)