原始数据为A股2009年至2019年的相关数据。我想要进行PSM并生成新的两个变量，TREAT和TIME。
在2009年至2019年期间，stkcd对应的multi值发生了变化的公司为处理组。对于multi从0变1的处理组，其控制组为multi一直是0的公司；对于multi从1变0的处理组，其控制组为multi一直是1的公司。将处理组和控制组按照最邻近匹配方法（1:1）进行匹配，匹配变量为 roa lev。

匹配后的样本中，

TREAT为1，当stkcd对应的multi的值在2009-2019之间发生了变化（无论是从0变成1还是从1变成0）；TREAT 为0，当stkcd对应的multi的值在2009-2019之间一直维持不变。
TIME为1，当year为multi发生变化（无论是从0变成1还是从1变成0）当年及之后；TIME为0，当year为multi发生变化（无论是从0变成1还是从1变成0）之前。


lev multi roa stkcd year
-.9243 0 .046727 000002 2009
-1.3392 0 .040993 000002 2010
-4.1312 0 .03916 000002 2011
-.1422 0 .041348 000002 2012
-.4032 0 .038183 000002 2013
-.2545 1 .037937 000002 2014
-1.2667 1 .04245 000002 2015
-.5526 1 .034129 000002 2016
-.1614 1 .031929 000002 2017
-.3136 1 .032234 000002 2018
-.2241 1 .031869 000002 2019
-.8518 0 .04683 000004 2011
-.5756 0 .040506 000004 2012
-.9952 0 .023938 000004 2013
-.8703 0 .046189 000004 2014
-.4324 1 .0172 000004 2015
-1.1731 1 .17753 000004 2016
-.9061 0 .021917 000004 2017
-.8703 0 -.061543 000004 2018
-1.6953 0 -.002412 000004 2019
-.6448 0 -.037083 000005 2009
-.1276 0 -.003422 000005 2010
-.2966 0 .009055 000005 2011
-.1422 0 .016687 000005 2012
-.4032 0 -.035624 000005 2013
-.2545 0 .031583 000005 2014
-1.2667 0 -.024627 000005 2015
-.8839 0 .046281 000005 2016
-.3557 0 .007876 000005 2017
-.666 0 .050217 000005 2018
-.569 0 .058875 000005 2019
-.9243 0 .046098 000006 2009
-1.3392 1 .057942 000006 2010
-4.1312 1 .052165 000006 2011
-.1422 1 .067381 000006 2012
-.4032 1 .070024 000006 2013
-.2545 0 .043601 000006 2014
-1.2667 0 .034541 000006 2015
-.5526 0 .061037 000006 2016
-.1614 0 .062442 000006 2017
-.3136 0 .068194 000006 2018
-.2241 0 .053544 000006 2019
-.2407 0 .005614 000007 2012
.0206 0 .028639 000007 2013
.016 0 -.049025 000007 2014
.0014 0 .02756 000007 2015
8.4036 1 .158579 000007 2016
.0206 0 .036318 000008 2013
.016 0 .011522 000008 2014
-1.4164 0 .052909 000008 2015
-.4136 0 .064758 000008 2016
-.1846 0 .08331 000008 2017
-.1962 1 .032666 000008 2018
-.3019 1 .037977 000008 2019
-.2966 0 .03228 000009 2011
.0809 0 .023331 000009 2012
.1903 0 .03263 000009 2013
-.3518 0 .032091 000009 2014
.411 0 .053324 000009 2015
.479 0 .0194 000009 2016
-.4775 0 .011924 000009 2017
-.1001 0 .022281 000009 2018
-6.4791 0 .023564 000009 2019
-.6501 1 .003708 000010 2013
-.9716 1 -.093654 000010 2014
-.0957 0 .003869 000010 2015
-.7613 0 .008022 000010 2016
-.3462 0 -.33815 000010 2017
-.4595 0 -.221953 000010 2018
-.6448 0 .034199 000011 2009
-1.3392 0 .060069 000011 2010
-4.1312 0 .073569 000011 2011
-.1422 0 .094875 000011 2012
-.4032 0 .077671 000011 2013
-.2545 0 .107512 000011 2014
-1.2667 0 .035806 000011 2015
-.5526 0 .053327 000011 2016
-.1614 0 .115506 000011 2017
-.3136 0 .101828 000011 2018
-.2241 0 .068891 000011 2019
-.1681 0 .034983 000012 2015
-.6208 0 .046902 000012 2016
-1.061 0 .042418 000012 2017
-.5246 0 .024705 000012 2018
3.9013 0 .03079 000012 2019
-.9243 0 .047744 000014 2009
-1.3392 0 .048823 000014 2010
-4.1312 0 .035961 000014 2011
-.1422 0 .016419 000014 2012
-.4032 0 .018602 000014 2013
-.2545 0 .028069 000014 2014
-1.2667 1 .024607 000014 2015
-.5526 1 .015537 000014 2016
-.1614 0 .00719 000014 2017
-.3136 0 .080596 000014 2018
-.2241 0 .016291 000014 2019
-.3182 0 .010863 000016 2009
-.7971 0 .006145 000016 2010
-.991 0 .00143 000016 2011
-1.0384 0 .002743 000016 2012
-1.7197 0 .003801 000016 2013
-1.4919 0 .003607 000016 2014
-1.5313 0 -.089531 000016 2015
-.763 0 .005346 000016 2016
-.4495 0 .215921 000016 2017
-.1919 0 .02027 000016 2018
-.5607 0 .007865 000016 2019
.744 0 .112314 000017 2014
-1.4164 0 -.002309 000017 2015
-.4136 0 .071949 000017 2016
-.1846 0 .021468 000017 2017
-.1962 0 -.025675 000017 2018
-.3019 0 -.124557 000017 2019
-.8589 0 -.015826 000019 2009
3.8924 0 .013646 000019 2010
-1.0921 0 .009081 000019 2011
-1.5546 0 .060559 000019 2012
-.4446 0 .033827 000019 2013
-.4858 0 .012235 000019 2014
-.7167 0 -.040511 000019 2015
-.463 0 .076489 000019 2016
-.8571 0 -.057225 000019 2017
-.9617 0 .049599 000019 2018
-1.0103 0 .056755 000019 2019