Setting iniziale e visualizzazione effetto
Stabilisco i parametri del modello
intercept = 0
multitaskingEffect = -.28
dysfluencyEffect = .39
interactionEffect = .20
sigma = 1
Calcolo i livelli medi nelle varie condizioni per visualizzarli
# visualize effect
mean_MT0_DF0 = intercept
mean_MT0_DF1 = intercept + dysfluencyEffect
mean_MT1_DF0 = intercept + multitaskingEffect
mean_MT1_DF1 = intercept + dysfluencyEffect + multitaskingEffect + interactionEffect
Visualizzo con grafico ggplot
dataplot = data.frame(estimate = c(mean_MT0_DF0, mean_MT0_DF1, mean_MT1_DF0, mean_MT1_DF1),
multitasking = c("MT0","MT0","MT1","MT1"),
dysfluency = c("DF0","DF1","DF0","DF1"))
dataplot$dysfluency = factor(dataplot$dysfluency,levels=c("DF1","DF0"))
ggplot(data=dataplot,aes(y=estimate,x=multitasking,group=dysfluency,color=dysfluency,shape=dysfluency)) +
geom_hline(yintercept=0, linetype=2, size=0.8)+
geom_point(size=7) +
geom_line(size=1.5) +
theme(text=element_text(size=27)) +
scale_y_continuous(breaks=seq(-2,2,.1)) +
ylab("z-score (1 = SD)")
![](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABUAAAAPACAMAAADDuCPrAAABsFBMVEUAAAAAADoAAGYAOjoAOmYAOpAAZpAAZrYAv8QzMzM6AAA6OgA6Ojo6OmY6Zjo6ZmY6ZpA6ZrY6kLY6kNtNTU1NTW5NTY5Nbm5Nbo5NbqtNjshmAABmOgBmOjpmOmZmZgBmZjpmZmZmZpBmkJBmkLZmkNtmtttmtv9uTU1uTY5ubk1ubm5ubo5ujo5ujqtujshuq6tuq8huq+SOTU2Obk2Obm6Ojm6OjquOq6uOq8iOyOSOyP+QOgCQOjqQOmaQZgCQZjqQZmaQZpCQkDqQkGaQkJCQkLaQtraQttuQtv+Q27aQ29uQ2/+rbk2rjk2rjm6rjo6rq46rq8iryMiryOSr5P+2ZgC2Zjq2kDq2kGa2kJC2tpC2tra2ttu229u22/+2/9u2///Ijk3Ijm7Iq27Iq47Iq6vIyKvIyMjI5OTI5P/I///bkDrbkGbbtmbbtpDbtrbbttvb25Db27bb29vb2//b/9vb///kq27kyI7kyKvkyMjk5Mjk5OTk5P/k///r6+vy8vL4dm3/tmb/yI7/25D/27b/29v/5Kv/5Mj/5OT//7b//8j//9v//+T///+ATlkZAAAACXBIWXMAAB2HAAAdhwGP5fFlAAAgAElEQVR4nO29+2McVZbnKZXF4LLlxS6SwrjsXQxu4yl144WiGgrYXbPLgneHaqpnXKJ5GLZEu2fapbE9a7HI3fS4QLIMBjn/5VVmPDLivuLeiBNxThx9Pz/MlDNvpvKcvufDfUTcWJoCAABoxRL3DwAAgLECgQIAQEsgUAAAaAkECgAALYFAAQCgJRAoAAC0BAIFAICWQKAAANASCBQAAFoCgQIAQEsgUAAAaAkECgAALYFAAQCgJYoE+hcVaIlDAFpSyV1YIAAEKgwtcQhASyq5CwsEgECFoSUOAWhJJXdhgQAQqDC0xCEALankLiwQAAIVhpY4BKAlldyFBQJAoMLQEocAtKSSu7BAAAhUGFriEICWVHIXFgjAJdBvPlybHLD2xs20z20ffOi5+863uPs5DVriEICWVBKUG+gLHoHeuzQpOX0t4YO7EwgURKIllTQ1B3qBQ6CPfz+pcTn6kz/9BgIFsWhJJVHZgT5gEKjpz8nkTOxHNyYQKIhFSyrJKg/QwyDQDdOfk8mrcZ/cnkCgIBotqSSsPUDN8ALdLZY+DzT46Ea2GPrMH2M++f0lCBTEoyWVpOUHaBlcoNkyZjnmfPxlyImuT0KgIA4tqaQrPkDO4ALdNufs2Yz+reZPllN/CBTEoCWVRJUH+mBogWY7SGeCr7iZm3dNu0C/++477p+gBiVdAgKVzNACna9j1pc854uiv/xzzAfPaL4O9Lsq3D9GA+PvEhlUtQd6YGiBbtsGzNY2G+bw83HqL/+sV6DfmXD/oPEz9i5RQFZ8gJ6BBZrN142rljYi5vAbmWS1CtTSJxTanXF3iQVEtQf6YGCBOkebIStWm7yq9lZOpz9h0I6MuktUIKs+QM/AAnUsgeYvBhdB5y1m3lQpUI8+odCOjLhL1KArP0DOwAJ1bhg5rVplPvGfN9Ao0IA/YdAujLdL1KErP0AOh0BNAc7n9SGBbpfzfoUCDfoTBu3AaLuEAVn1AXoGFui2a78o21nyb8PvLj4EgYJoRtslDKiKD/TACAQ6H6Dm036XQH+eQ/5jB6HBn999x/0DAQBe5At0sQA6VSjQRn/CoADIZWCBOi/5DAt0u3rlqLopfIRAuX/iaBlpl7CgKj7QA+IFulv7hDaBRvgTBm3LOLuEDVXxgR6QLtDqAugUAgUJjLNL2JBVH6BH+hroRv0tZQKN8icM2pJRdgkHVMUHekC4QGsLoFMIFCQwyi7hgKr4QA/IFmh5C2cBBAqiGWWXcEBWfYAe0Xci2W9AoCCaUXYJB3TlB8iRIFDvvfD58+fcWCNW7n7eAgi0T0bZJRz0VYyAAI7TmJyHibhOY9Iu0Eh/wqDtGGOXcNFXMQICOM4DNQebjlPqMyBQCLQDY+wSLvoqRkCA6BPpIVAItANj7BIueqpFQIHoZyJpFyjWQHtllF3CQW/FCLozmqdyVtoeul14SLQVo+wSDrpXHeiN8TwXfs7hFSgkmswou4SD7lUHemNogWZz+Ooq6IZ7Qu7mcAsUEk1ilF3CAUHVgb4YXKDZiufklVyCjzN/hp/JuQAChUSjGWWXcEBUeaAPBhdoPgSdPPvmgQYf3bg0/0fogUg1lAm0rUHh0BjG2SVs6GoPkDO8QLM1zzqvmm/79pQgUEg0mnF2CZseihBQMbxApz+9ZvrT3lM6LALtZlBINMhIu4QFfQkCMhgEao1BL9tvQqB1R0KiLRhpl7Cgrj9ACIdAp9N7lxb6PH2t9tYhE2izQeNaMkYglrF2CRPq6gOE8Ah0Ov3mw7WZC9dev2m8cdgEmvZceEg0gdF2CQPaygOkcAm0B7j7eUuSBNr4geF/v1xG2yUMuAsLBIBA2WkhxFTnHk7G2yXqcBcWCACB8tPOhZBoEyPuEjW4CwsEgEAl0NaCkGiIUXeJCtyFBQJAoCLoIEA41Me4u8QC7sICASBQIXRyHyTqYuxdooC7sEAACFQMHa0HiZqMv0tkcBcWCACByqKb7CDRKkq6BAQqGQhUGJ3jgEQL9HQJIBYIVBgkcUCiM/R0CSAWCFQYZHFAonq6BBALBCoMyjiCDtUvUT1dAogFAhUGdRyHWKJ6ugQQCwQqjD7iOKQO1dMlgFggUGH0FcchlKieLgHEAoEKo884DplE9XQJIBYIVBh9x3GIJKqnSwCxQKDCGCKOQyJRPV0CiAUCFcZQcRwCierpEkAsEKgwhoxDuUT1dAkgFghUGAPHEXToyCWqp0sAsUCgwmCIQ6tE9XQJIBYIVBhMcWiUqJ4uAcQCgQqDMQ5tDtXTJYBYIFBhMMehSaJ6ugQQCwQqDAFxaJGogFSSwF1YIAAEKgwhcWiQqJBUdoa7sEAACFQYguIYu0QFpbIT3IUFAkCgwhAWx5glKiyVreEuLBAAAhWGwDjGKlGBqWwFd2GBABCoMGTGEXSoVInKTGU63IUFAigSKOiXsES5fx0AHCgSKPdAgQbZcYxqICo7lfFwFxYIAIEKQ34co5Go/FTGwV1YIAAEKoxxxDEKiY4jlc1wFxYIAIEKYzxxiJfoeFIZhruwQAAIVBjjikO0RMeVSj/chQUCQKDCGF8cYiU6vlS64S4sEAACFcY44xAp0XGm0oa7sEAACFQYo40j6FAWiY42lQbchQUCQKDCGHUcsiQ66lRW4C4sEAACFcbo45Aj0dGnMoe7sEAACFQYKuKQ4VAVqfwLBCoaCFQYWuIQIFEtqeQuLBAAAhWGljjm8EpUSyq5CwsEgECFoSWOEj6Jakkld2GBABCoMLTEUYNHolpSyV1YIAAEKgwtcVgML1EtqeQuLBAAAhWGljicDCtRLankLiwQAAIVhpY4fAQdSitRLankLiwQAAIVhpY4QgwkUS2p5C4sEAACFYaWOJoYwKFaUsldWCAABCoMLXHE0K9E+1hWZYG7sEAACFQYWuKIpReJ9reuygF3YYEAEKgwtMSRArFEe1oRYIO7sEAACFQYWuJIhUyi/awHcMJdWCAABCoMLXG0gUKiZONYOXAXFggAgQpDSxxt6SZRolGsLLgLCwSAQIWhJY4OBB0aFCHFEFYe3IUFAkCgwtASR0daSTT8odEalLuwQAAIVBha4iAgWYcQKBgaCFQYWuIgIsWIDf4crUG5CwsEgECFoSUOQiKl2OjPsRqUu7BAAAhUGFriICbCixAoGB4IVBha4uiBsBsj/DlSg3IXFggAgQpDSxw9EaVJCBQMBZdAv/lwbXLA2hs3o5r/8Pfz5s++fO2+tw13P6dBSxw9ctgMSlVzoAd4BHrv0qTk9LXG5j+8tmg+uexTKHc/p0FLHD0DgQIRcAj08e8nNS43tN+uN3/mTXcz7n5Og5Y4+gcCBfwwCNT052RyJth+w2w+edXZjruf06AljmGAQAEvDAK1hegxYsa23Xzylqshdz+nQUscwwGBAj6GF+husfR5fzp9dCNbDH3mj97m3+ctfvffD/7xb19m//rlnx0tufs5DVriGBbVBu2tFEF3BhfoT7+pjTkffzn/53O+naF8vl++n//bNWTl7uc0aIljYCBQwMLgAt02Bbjhn5RPi/Fqxa+ZQV3C5e7nNGiJY2AgUMDC0ALN/Hcm+EqVDWuGv+ub83P3cxq0xDE0ev0JgUpmaIHOlzTr+tv1rmo67TpfA4BAQR0IFHAwtEC37Ql4tirqnsNnW0hvWc0hUFAHAgUcDCzQbERpbAFtBObwj//173/772ujUwgUOIBAAQcDC9Q52twN7sObeFtz93MatMQxNBAo4GBggTqWQPMXPYugFpmCcRkTMFDrTwhUMgML1Llh5LSqh3+55B2ucvdzGrTEMTgQKGCAQ6Cm/ryrmib/duPX/huRlIhHSxzDo9WfEKhkBhbotmu/KNtZ8l1Kn5PfwRRYLuXu5zRoiWN4IFAwPKMR6PfFEaLWaXY/z6H8nWCENPqT+wcCfYxGoLvlAPQZ40hlCBRkwJ9gaJIEurV157MDrm991fbPOS/5TBSo70hl7pkWDVri4EDlBB5TeNFECvThp++sLtVYfeHq1+l/rr1At//69f/0X/7wWq5QXMYEHKj0JwQqmQiB7t8+t+Rm+eTVxD/XXqBF2y+9BuXu5zRoiYMHjf6EQCXTKNDbZz32zDnxecqfa78GWnJv4jmOibuf06AlDi7U6RMCFU1YoA8/OhbW53wceiF+Lk8g0PxEUfveee5+ToOWONhQ508IVDIhge5dabZnxqnYXSUKgWbN7Yvpufs5DVriYESXPiFQ0fgFGq/PGSe/iPpz3e5Eqn6J3Z67n9OgJQ5WNOkTAhWNT6D7Hxnz9BPn3/9sK7t+6eHW1mefnDd25ZdOxUzknQJNuRe+bG+PWLn7OQ1a4uBHgzvnRBcGGB6PQG9V1z5PXHBvFO3feed41bEXm/+c8+ClpNOYpt4Dmbj7OQ1a4hCAllRGFwYYHqdA9xY778snw7vstWucVhrn8c7ZuuOU+pJH//kPvzbXTCFQEIOWVDYVFWDEJdD/Wg4/41Y2K1c6/aJhHp96Iv229/g7TOFBA1pSGVODgAlboHtv5zI8cjH68qTF5U5Ng9DEZyJlN3C6TrDHJhJoQEsqY6sQMGAJtFj9XEm7yWj/1vGoldDEp3Jmg8366BSXMYEotKQyqRDBsJgC/fFXbfQ5J5/J/+xPoUaJz4XP3nQIFxfSgya0pDK9FMFgOAUas5/uYj56DQs0v4+osgq64ZqlG82rw83dS7iVE8SgJZUtKhEMhUugUVd0Otn/+FiTQPOT5V/Jl0EfZ/70PpMzb376ZvFCZlTXiJW7n9OgJQ4BaEll22oEA2ALNPKeIg97VxoEWhjw2TcPnPnoxiXf0SAFxTmgL//zdNY+eyiSc8mUu5/ToCUOAWhJZZd6BD1jCnT/v3X9xv8WdSVTnVfNtyuC3LCbu4XL3c9p0BKHALSksmtFgh4Z+JEeM356zfShvadUHWF+GedPNdXC/QvUoCWVPdcj6AKDQK0x6GX7zdoUfdvw52JBtAZ3P6dBSxwC0JLKHksRdIVDoNPpvUsVHV6rveW6yvOH31b0+azzgUhTPdXC/QvUoCWVfVUhIIBHoNPpNx+uzWy49ro5mnRfJv/4n347vwJ/7Q336HMGdz+nQUscAtCSSvrqA2RwCbQHuPs5DVriEICWVHIXFggAgQpDSxwC0JJK7sICASBQYWiJQwBaUsldWCBArEC3ZrS+Q2kQuPs5DVriEICWVHIXFgjQKND9O++cqxxPv/qU53h6frj7OQ1a4hCAllRyFxYIEBbo3rvOxxqvXJU4FuXu5zRoiUMAWlLJXVggQEigt4677JlyVv2gcPdzGrTEIQAtqeQuLBDAL9DKgzrGoVDufk6DljgEoCWV3IUFAsQ+1tjB8l8N+ksb4e7nNGiJQwBaUsldWCCAR6B7DcPPjCdELYVy93MatMQhAC2p5C4sEMAt0Af23tGR1VXboCsNZ38OCnc/p0FLHALQkkruwgIBnAKt+fPEhetb5TsP7376fPXNptOTh4S7n9OgJQ4BaEkld2GBAC6B5g+Wmw8xXQ+Xmz24Q+Asnruf06AlDgFoSSV3YYEALoGul/r0brQvtuif7vHHpcHdz2nQEocAtKSSu7BAAIdAdwo3/iIwutz/h2Iv/oP+flwa3P2cBi1xCEBLKrkLCwSwBbr/dm7GhmcbFwulT/b0y5Lh7uc0aIlDAFpSyV1YIIAt0GIA+lLTR3ODihmCcvdzGrTEIQAtqeQuLBDAFuh69NrmpqxVUO5+ToOWOASgJZXchQUCWALNt+Bjdtfzyb6UjXjufk6DljgEoCWV3IUFAlgC3YmcwC/aSpnDc/dzGrTEIQAtqeQuLBDAEuhmwqgyH61GybZ/uPs5DVriEICWVHIXFghgCXQ9ZWt9XdIiKHc/p0FLHALQkkruwgIBPAKNdOKmpAuZuPs5DVriEICWVHIXFghgCjTfGIqcle9I2kXi7uc0aIlDAFpSyV1YIIBboLH7QtmloBAoIVriEICWVHIXFghgCjTbF0oTqJAjmbj7OQ1a4hCAllRyFxYI4BZorBLTWvcMdz+nQUscAtCSSu7CAgEgUGFoiUMAWlLJXVggAAQqDC1xCEBLKrkLCwSAQIWhJQ4BaEkld2GBAIoECgAAw6JIoNwDBRq0xCEALankLiwQAAIVhpY4BKAlldyFBQJAoMLQEocAtKSSu7BAAAhUGFriEICWVHIXFggAgQpDSxwC0JJK7sICASBQYWiJQwBaUsldWCAABCoMLXEIQEsquQsLBIBAhaElDgFoSSV3YYEAntOY3v8sik9xGhM1WuIQgJZUchcWCOAWaBoQKCFa4hCAllRyFxYIAIEKQ0scAtCSSu7CAgEgUGFoiUMAWlLJXVggAAQqDC1xCEBLKrkLCwSAQIWhJQ4BaEkld2GBABCoMLTEIQAtqeQuLBDAeirnnbgLmKpcx1M56dAShwC0pJK7sEAAU6Ajhruf06AlDgFoSSV3YYEAEKgwtMQhAC2p5C4sEAACFYaWOASgJZXchQUCQKDC0BKHALSkkruwQAAIVBha4hCAllRyFxYIAIEKQ0scAtCSSu7CAgEgUGFoiUMAWlLJXVggAAQqDC1xCEBLKrkLCwRoJdCtra+ofwcB3P2cBi1xCEBLKrkLCwRIFujtK/NDlJeOvPhFH7+nA9z9nAYtcQhASyq5CwsECAh0/87HL5g3ad46VrkHfuVqrz8tFe5+ToOWOASgJZXchQUC+AS6f+u4fUzI/hXjGJFTMu6Cz+Du5zRoiUMAWlLJXVgggEeg+UizLtD9t62DmFYEzeO5+zkNWuIQgJZUchcWCOAUaDnSrAnU4U8xR9nN4O7nNGiJQwBaUsldWCCAS6ALU9b0uO48DPQJMbN47n5Og5Y4BKAlldyFBQI4BFoZaVYF+qDcPzry4meffXy2+NeTw/3YMNz9nAYtcQhASyq5CwsEcAh00zlBL7W6nO+97+UvLH8w0E9tgruf06AlDgFoSSV3YYEAtkDLkebyherl8ju2UzdlTeK5+zkNWuIQgJZUchcWCGALNF/qXL7ofvkD67VWQ9BvPlybHLD2xs2o5o//6W9/PWv/7N/8x/u+Ntz9nAYtcQhASypblBcYCkug+QDUtGIxLn26+mL+BLr0VdB7lyYlp681Nv/htUmFv/Y4l7uf06AlDgFoSWVyeYHhsASaz8tfcr9sXLW02epSpse/n9S43ND8y4nBq8523P2cBi1xCEBLKtOqCwyK9VTOt52DymILyXg9G5cmzuFNf04mZ9Kaez7A3c9p0BKHALSkMqm6wLCYAvUosXhcvDEwzb369DSFDVuI7iGlt7n7A9z9nAYtcQhASyqTqgsMiynQHffG+o5zBl/M4ZMWQXeLpc/70+mjG9li6DN/bGr+zMv/fPCPx998mBv0Lbsldz+nQUscAtCSypTqAgNjCnTTPaRc91w07/FtgJ9+UxtC5gucz/n21vPmi/fzDSXHB7j7OQ1a4hCAllTGFxcYHFOg684tpGIJ1Hw9n/GnCHTbnIJv+EaUc7IB6JmKLvM1UfsD3P2cBi1xCEBLKuOLCwyOIdDMlNYSqOfapmJtNGEbPtPfmeArVeZ6/eWfqy99f8m9Csrdz2nQEocAtKQyurjA8BgC9Qhxx3fPUbJA5/arL3nu2o4syWbwhiw33HN47n5Og5Y4BKAlldHFBYYnTqC+JdB0gW7b8ssk6Z7DO3ybGxcCBQ1oSWV0cYHhiRKodwk0WaDZfN01onTP4Z2jU8+Qlbuf06AlDgFoSWVscQEGogSaXwXquGA+WxyNF6hztOkZUc559K9/+K35FkagIAYtqYwtLsBAlEC9S6DJu/DOKfn8Rc8iqItt94iVu5/ToCUOAWhJZXRhgOGJ2oX3LoEmX0jvnH07rerHuQww1VMt3L9ADVpSGV1dYHhirgP1L4HmzRMFas6+5/P6eIHueu5d4u7nNGiJQwBaUhldXWB43AKt34n0wPWIzjn54mj8vfDO2Xc2pPRdSm+SLaPiTiTQhJZURlcXGJ6Ye+H9J8/v+DaXfBAI1HvnEnc/p0FLHALQksro6gLDE3EaUzGDt8eZ+Tupl4F2Emh2K2j1K36eE/0jAACAhIjzQPM9eMc4c8e7ueTDeclnikAzf9a2oSBQAAAPzSfSe85SngY3l3x0Feh24PQ77pkWDVriEICWVMaXFxgcS6DF0cnFM+X282uYHJpc966Neukm0OLhHu6m3P2cBi1xCEBLKuPLCwyO/VTO4rHwJ7+Y/evu2SWPJkuzppxH32kNtHi4h+f8eu5+ToOWOASgJZUJ9QWGxhZoMQRdWlpeXV0q/7e5Anrba9YQXQRaPJvzTc/73P2cBi1xCEBLKhPqCwyNLdBia6hOfQV0v9Rn4hPlOgj0+/zpH9523P2cBi1xCEBLKlMKDAyMQ6DlJL5CdZj54/9yrPJO2gPl2t+JlG0fhZpx93MatMQhAC2pTKowMCwugdoGrU3TF3P8ZH+6BRpxL3yxfXT6pr8Ndz+nQUscAtCSyrQSA4PiFOj0wdmaP0/VljmrAk30p/vgpebTmIrlT++z52Zw93MatMQhAC2pTKwxMCRugU73by2m6Suf19+r7DJdTf1zztm645T6Ovny5+Ry8Lu5+zkNWuIQgJZUphYZGBCPQA/Y+/j51dXVp97zPAfpQJ+nUvbfM1JPpJ9TbB/5tt9zuPs5DVriEICWVCZXGRgOv0C9ZAI9ciFdn9PkZyLNKPzZdN4ddz+nQUscAtCSyjZ1BgaijUDPnX/xeit7TpOfyjkt/Xm68cR67n5Og5Y4BKAllS1LDQxBC4F2IvW58MXtR8Htowzufk6DljgEoCWVFHUHemJogebXc1ZWQb3ne1bfPtPsTzXVwv0L1KAllRRlB3picIFmK56TV3IjPt4IDzB3Y8efeqqF+xeoQUsqqUoP9MDgAi1uKXr2zQMpPrrRsEGUTeDjntjJ3c9p0BKHALSkkrD4ADXDC7Q4U6nKq+bbhTJ37bY5tlS5+zkNWuIQgJZUDlWZoAXDC3T602umDO09pUKPGxAoaIuWVA5Ul6ANDAK1xqCX7TdzPeYLphAoaIGWVA5TlaAVHAKdTu9dWpjw9LXaWzWBfn/J508IFDShJZXD1CRoBY9Ap9NvPlybaXDtdfN4pZpA/UugEChoQksqh6hH0BIugfYAdz+nQUscAtCSSu7CAgEgUGFoiUMAWlLJXVggAAQqDC1xCEBLKrkLCwSAQIWhJQ4BaEkld2GBABCoMLTEIQAtqeQuLBAAAhWGljgEoCWV3IUFAkCgwtAShwC0pJK7sEAACFQYWuIQgJZUchcWCACBCkNLHALQkkruwgIBIFBhaIlDAFpSyV1YIAAEKgwtcQhASyq5CwsEgECFoSUOAWhJJXdhgQAQqDC0xCEALankLiwQAAIVhpY4BKAlldyFBQJAoMLQEocAtKSSu7BAAAhUGFriEICWVHIXFggAgQpDSxwC0JJK7sICASBQYWiJQwBaUsldWCBAg0B//NXSAT/7k+/fkuDu5zRoiUMAWlLJXVggAAQqDC1xCEBLKrkLCwSAQIWhJQ4BaEkld2GBABCoMLTEIQAtqeQuLBAAAhWGljgEoCWV3IUFAkCgwtAShwC0pJK7sEAACFQYWuIQgJZUchcWCACBCkNLHALQkkruwgIBIFBhaIlDAFpSyV1YIAAEKgwtcQhASyq5CwsEgECFoSUOAWhJJXdhgQAQqDC0xCEALankLiwQAAIVhpY4BKAlldyFBQIoEigAAAyLIoFyDxRo0BKHALSkkruwQAAIVBha4hCAllRyFxYIAIEKQ0scAtCSSu7CiuT2ldWZHJZXz7/3NfdvGQ4IVBha4hCAllRSFcjmUo/Fe+vYUsnBn9jp829JAgIVhpY4BKAllVQF0qdAN5cqPPE1BJoDgQ6NljgEoCWVVAXSo0Br/lx6cgqB5kCgQ6MlDgFoSSVVgfQn0AfHagJ9GgItgECHRkscAtCSSqoC6U+g6zV/Lr0EgRZAoEOjJQ4BaEklVYH0JtD9tzNxLl/4avavra8h0AIIdGi0xCEALamkKpDeBJpZYWn5g8VLEGgGBDo0WuIQgJZUUhVIbwLNl0CfrrwEgWZAoEOjJQ4BaEklVYH0LdCXKi9BoBkQ6NBoiUMAWlJJVSC9CXTHmsFDoDkQ6NBoiUMAWlJJVSAQaA9AoMLQEocAtKSSqkAg0B6AQIWhJQ4BaEklVYFAoD0AgQpDSxwC0JLK1hWx9+7qbHtn9YUv5v9sL9CHn5w7Pnfk6gtXHSct9SLQ4m8eOfHeVw1N8ziPnLgQaPnw4/z7niI9LAoCFYaWOASgJZUt66F6PtLKTKFVgebXvrsqOb+qs9xT3791fKnKytVF2+IS+irzD5oC9Qk1u4fpCUtp9b9Z/ZPV73vaamq3zGM45v6+hDy4gUCFoSUOAWhJZatqeHC2rrWnjRHozpJPD4brzC8qdDynJ4Hetv7m4k9Wv+8gqD2z6UnH+LKuz/lf/KL2RRF58ACBCkNLHALQkso2xXDL8tqpukDz8dWT9kfXa2/YX3TAcuGbfgS6aX/p0vLFWpNCoA8sNS6tmH/E9SPLBYfYPPhoEOiY4O7nNGiJQwBaUtmiFlwKerq+BrruUVpmlMIvOy5/LpUjtj4Eur9uf+f891c/lwvU4U9Lxz/aQ+hqhHF58AKBCkNLHALQksr0UnBrb7WmCt/cdacqoXx8ZpN/Sx8C9fjTdavoL1xjSyMm5/iz8mOi8uAHAhWGljgEoCWVyZWw0N7Jzw8EsH/7nO0N39w1900uq0JmJ97fmv3z4e0rdZv1INBy8Lz84uzHT7c+LkaZ9mEl+etmlG4fZ9+3f7cM4cn4PPiBQIWhJQ4BaEllciUUzih3SqpbLaXENi0rzcgmxfmrjmOWim+qSiriMqZogRaT8uqaZ7EHVPl4RaC/+Nr8bHU4WX7fhbJZEVwqpf4AACAASURBVEL+gyPyEAACFYaWOASgJZWphVA441TltcXC4mJ73T5DaVroJDeaw4z+s+tIBFoMaeu77oXyFr+1FGjtrxaRL9qVR5VWmxUvPhmbhwAQqDC0xCEALalMLYT1qh7MVxcSyzVSd0R95rrp+iJbOIQC3VlyNsytbX7fkrl2mb/8hDkoNUaSta+LyEMACFQYWuIQgJZUJtaBJZvay5XXXXPXTDj1nXrDIrlbKlolFOi6y3dTe6unEKghdyv2dXNI6vi65jwEgECFoSUOAWhJZWId7LidYQ/uXHPX+pjTfSnk5pETL17fsr6ZQqD5T/JelVk2zIOxTGtclJQL1ZqJ13aOmvMQAAIVhpY4BKAllYl14Lvl3RqdOeau+UvFQM83HjSgE6hr0dX5BdZcPccI3r2+Ofu7R5568fpXtaADeQgAgQpDSxwC0JLKtDKwZ9gF1iXjtq6MmWt+SZFxF5AFnUB9t8Zbd6bnArWiNATq3mFv/v3RM3gIVBpa4hCAllSmlUFuGsf2hyUxu6kxcy2vC1o+eT3wJ8kE6pe/uasTXqcof4pfyFUa8xAAAhWGljgEoCWVaWXgeEBR7Z2qxDwXEb1k/DvjyIufe/4kmUD98jfXY8P3DxU/JSDkwK9ImcFDoNLQEocAtKQyrQz8y4j2SUBmW+v2RfNe8xOu40DpBOq8tb1G0dIXZf31kJBDAUTexjkDAhWGljgEoCWVaWXgP4UtG1dV3zEFY1+2ZJ9KcuSC5RUygfqOLuko0MaRZHMevHAJ9JsP1yYHrL1xM+VT31+aTF71vcndz2nQEocAtKQyrbL8587bAjXut3EdQOQ618k8tXhAgdaPQmkSqH85w6A5Dz54BHrv0qTk9LX4z21MIFAQi5ZUptVW4MEd69Y79Tu+nTNX+zDiJfNOy/ELNCIPHjgE+vj3kxqXYz+4PYFAQTRaUplWXUkCdZ29ZM5czUd6ZIqqSmn8Ao3JgxsGgZr+nEzOxH1wdwKBgni0pDKtvJIEWpu7+meuDz+2hqG9HCbCJtDIPDhgEOiG6c+AFKvsXoJAQQJaUplWXimbSPW5a3Dmuv+JcbB7pR2xQGPURbuJlJAHk+EFulssfd6fTh/dyKT4zB+bP3evSbbc/ZwGLXEIQEsq0+rLLyGXQKtz1/Um29x9tzqZXzQkvoyJXKARk/GkPFQZXKA//aamwcdfzv/53P2Gjz3+x8bRKnc/p0FLHALQksq0AvNPW51PhFwMt+IeGOk69536OtAId8UJNPZC+ml6HgoGF+i2acFsRv9W+FM/LNZNIVAQh5ZUphWYf9TlvMF7seC3Eymbvfz2JNOOEQK1hGec4+F/RKZJnED9t3L++KvV8+9//nX1hdQ8ZAwt0GwH6UzwFftDNyqXPUGgIA4tqUyrMOf5wHPc20vllnP0zDX/C+aNO20Eahw35//tJpEC9e6o2YsF6XmYM7RAZ5fCG0ue80XRX/7Z/5l/eS1T53N/B4GCeLSkMrHEfCcQ5XYybVLMXV0z1607Hz9/3NaZ+87HCIFaWjJPpdt0N5vOVHvkqQuLU0gjBeq7Zd5h1lAeAgwt0G17xTNbFfXP4Ytd+8v3NyBQEI+WVCaWmO9Q4vx190n1yx84Zq6+R6bTCXTdEGj+Gx1DUOP8ukiB+g5Udgx1A3kIMbBAs/m64cCN8Bx+o7xhCQIFCWhJZWKNWTPs2su2D/Mp66b9oXw86DkzLmkN1G314jEjpciKH2kNGYuW9fNAGwVaGNoz8q1F5s9DiIEF6hxt7ob34WfafOZ396cQKEhCSypTi2zHacripnZLoJnb/t3b9lDNM2a1hm8RAnU8y3Naef5y+VXFEc7uR3WYJ9I3C9SdC+ev8echxMACdSyB5i/6F0EPtPnyzeJ/QaAgFi2pTC0y56OBy0NBLIFWD/182vmOoRNrYBohUNdXLR61vHi1GGkaavzI/H2xAi3H3dVc1B9r3JyHEAML1Llh5LTqgo2XixObIFCQgJZUJldZcUfkcnlq0v5HpRvsJc3FgUumkIovqqp4/x+sthECLf/IwpV7i1ubrLuaDji1eK1s2birb79eHDFaeSxJ8XXmp/15CMAhUHO2Pp/Xx9yMBIGCBLSkMr3MyqHdytXZg9PuXqncym4LdHGMsTlzXQzKVt6bb4A/vPtu0TjtscbVp4PMDrZ/eDuz2MpZ8++Wv33p5PWv5y3LS/fDNz95Xi+1eOTCVv3rzGGmPw8BBhbotmu/KNtZariUfgYEChLQksr0Mqs9iqPkF45bOeutrZmr/4j40MH2U4dAK2KssPwfrCVH9283f168QL3f59uad+XBDwQqDC1xCEBLKlvUWbGUWBPG/+cRaDltdrzlOk/ZY8sGgbp+0tJL+/aezY9nHQ1nnAr/Sd/rHoOuJOXBiwKB/jyH6CcCMH5sa/zsT67DROYUbnNd/Og2aN1cUQJ1jGaXL04dAp3uX3H+zdqwMEGgboMGHp0cfxHodHCBOi/5hEABIKbY66kMuLwC9V0tOeeBY0B4qm6fOIFWd42yn/TB1ClQ5yn4K3VXpgjU8X3L9oOdmvLgQYFAC7hnWjRoiUMAWlLZstb2rhjC8AvUf4bo1HEi/ckvnB9vFOiByI4bP8kj0Om+cYLzkYtGgzSBTvc/qn2fFUFMHtxAoMLQEocAtKSydbXt335+Zqxl98OIKzTdvrj/6fnV7PSN1Reut/45M/benX/RkaeuNrV8+Mnzq/OftfrU+/Gb4n7uvntuHsLqSf+fTryNc4aCNdAC7n5Og5Y4BKAllR1rLoK0A4j00iIPpkB/PLeazIn4IS8E2oSWOASgJZXR1dWWbPMk4eJHpbTJgyVQ18UGDSSsGUCgTWiJQwBaUhldXW3JZq4JFz8qpU0eBhYo7kRqQkscAtCSyujqakm2kZNw+6JSWuVBgkAb7oVfAIGCBLSkMrq6WpJdoJm0daKSVnkwBVo5daAPgToPXmo4jWkBBAoS0JLK6OpqCbaQMlrlwd6FNy93JRWoc7buOKXeDQQKEtCSyujqakd2rxG2kNrlwXEZUzkIXTkfxwvxf7PNifRGQwgUxKElldHVlcad2WFH+/mpSId4ANotD87rQIv7X3tIaotnIi2AQEECWlLZuejc7CxVOMQD0G55cF9I7ztYvzttnspZAoGCBLSksnPRuakd7nF4B6Ad8+C5E2mzr/8utXoufAEEChLQksrOReemesHNYb4GtFsefLdy+h7I15n5HL5qweyhmzEzeAgUpKAlld2LzknllLdTza310i0P3nvh13uaxGcrnpNX8mXQx5k/Y/bgIVCQhJZU0hSeTXFG/PJf9fUXxkGnPHgF6nskfWeyIejk2TcPnPnoxqX5P2Kuop9CoCAJLakkKz2DvXdmZzWtuo/GPER0yoP/NCbXs+cpyNY867xqvu3eU4JAQQJaUkldgYCQwHF22TJo0umiUfz0mulPe08JAgWd0ZJK6gIEhAQEmk/i6feRzDHoZftNCBR0RksqicsPUBI6ULnFAfeR3Lu00Ofpa7W3IFDuX6AGLakkLj5ASUig+f5+L9eIffPh2syGa6/fNN6AQLl/gRq0pJK08AAtwUd67IzqLi/ufk6DljgEoCWV3IUFAoSfifRwa8ZAP6Ur3P2cBi1xCEBLKrkLCwQY+KFyfcLdz2nQEocAtKSSu7BAAAhUGFriEICWVHIXFggAgQpDSxwC0JJK7sICASBQYWiJQwBaUsldWCAABCoMLXEIQEsquQsLBIBAhaElDgFoSSV3YYEAEKgwtMQhAC2p5C4sEAACFYaWOASgJZXchQUCQKDC0BKHALSkkruwQABToPtfdf3Gh1x3fnL3cxq0xCEALalkqicQgynQH391qpMA9z/q4/SmKLj7OQ1a4hCAllQy1ROIwRbo0vLF9l9361gvx99Fwd3PadAShwC0pJKpnkAMDoEuLa1cbfdld8/2dH5oFNz9nAYtcQhASyqZ6gnE4BTo0tKJL9K/au9cbwcwR8Hdz2nQEocAtKSSqZ5ADNYu/H89lj1M7mSiQuejz97OX46Bu5/ToCUOAWhJJVdBgQjsy5j2ChOe+Dz6W/ZvHS+erUz+IPlouPs5DVriEICWVLJVFGjGcR3o/ke5DJeOxD0qee9K8YGlblv43eDu5zRoiUMAWlLJV1KgEeeF9LfPlkZcea9BiXtXjpWNl1tuPtHA3c9p0BKHALSkkrOoQAPuO5H2P15YcWn55HWPRB9+cm6pAufwc6qnWrh/gRq0pJK1qkAY362clWn5nNXz71/fKu9Serh155Pnj9dbpO46kcPdz2nQEocAtKSSua5ACP+98HfPLiWwEr/j1Bfc/ZwGLXEIQEsquQsLBAgdJnI7WqEC9KmnWrh/gRq0pDKlCL5tQ18VeRgIn8a0d65ZngIm7xnc/ZwGLXEIQEsqU4oAAh2YpuPsattJLo5c5N06WsDdz2nQEocAtKQypQgg0IGJOA/04cfHvfaMu1B0GLj7OQ1a4hCAllSmFAEEOjBxByrv33n3nDESXT5xQcLCZwXufk6DljgEoCWVKUUAgQ5Myon0W3c/y/m887HLPcDdz2nQEocAtKQypQgg0IHBIz2EoSUOAWhJZUoReCX53XffQaA9AIEKQ0scAtCSypQi8LhzAQRKiyKBAnDoadCnW6GxX75p7SIvr5544bqj5U7gup2XCMPlR5FAuQcKNGiJQwBaUplSBI36dCk09sttgWbYB1+mCvTBMcaDhDsBgQpDSxwC0JLKlCKI8adl0Ngv9wnUvpcmVaDrSxAoO9z9nAYtcQhASypTiiBGn5ZCY7/cL9Cl5boWEwU6/2IIlBnufk6DljgEoCWVKUUQ6c+6QWO/PCDQpaXas3zTBJq1hkCZ4e7nNGiJQwBaUplSBLH+rBk09svnAq09NHLr7uJI4KoY50qMfbzPzjEIVALc/ZwGLXEIQEsqU4pgYIHOKI4Orr6RItBbuYAhUGa4+zkNWuIQgJZUphRBtD+rBo39crdASwU+uXglXqD7/7AEgcqAu5/ToCUOAWhJZUoRxPuzYtDYL/cJtDDoYhIfLdC9t5cgUCFw93MatMQhAC2pTCkCHoHOL0NaWnqiPJstUqC1wzIhUGa4+zkNWuIQgJZUphRBgj8XBo39cr9Af/xV3ZhxAi0eefHE/waBCoC7n9OgJQ4BaEllShEwCTS/EunJ2j+bBLqeDz1Pfb0OgQqAu5/ToCUOAWhJZUoRpPizNGjslwcEmg1Byzl8gkBXrub/CwJlhruf06AlDgFoSWVKEXAJNHNgqcxogS7Pn2kBgUqAu5/ToCUOAWhJZUoRsAl0p7YPHyvQ/B56CFQC3P2cBi1xCEBLKlOKIMmfhUJjvzwk0AfHqhKME2h5BAkEKgHufk6DljgEoCWVKUXAJtBsEbTYRUq6lRMClQF3P6dBSxwC0JLKlCJgE+j+29VdJAh0hHD3cxq0xCEALalMKQJJAnXhPo4eApUAdz+nQUscAtCSypQiYBNoJsHiTQh0hHD3cxq0xCEALalMKQIIdGAgUGFoiUMAWlKZUgQQ6MBAoMKYHuX+BWpQ0yUSYF4DhUBHDHc/p+HoURiUCC1dIqUI2C6kxy786OHu5zRAoGRo6RIpRcAm0PrN8BCozf6dd84dX2Twx//1am8/qh3c/ZyEo0dhUCqUdIlRnMaEC+nD7N86Xi5jZBl8cCw/CEAM3P2cBAiUDiVdYhwCbXEr5wLtAq0dHJ1nMEvRxaaPDgh3P6fg6FEYlAwdXWIUJ9IXjzxOOkxkgXKBPqjqs8hg/ozoUz3/vgS4+zkBRyFQQlR0ib+MRKC1q5gg0CqbxpUIWZaK46SfbPr4YHD3cwKOwqCEqOgSfxnBUzmn7Q5UXqBaoKY/swzul0/TExM3dz/vztGjMCghGrrEjJQi+DbaoN+SCnSz7gIItKTizyP/w7Eyg9l/cWrrHtxw9/PuQKCkaOgSM1KKgEmgmQ4Wb0GgBeX654nPjTTdKt55QshePHc/78zRozAoJQq6xJyUIvg21qDfkgp03VjOg0Bzipn6yvzs6Pp/Z/Y/kjUE5e7nnYFAaVHQJeakFMG3kQb9llSg2Z2bFWFCoDk7tUGmOVD/SNQQlLufd+XoURiUlPF3iYyUIvj22xiFGq1iv9wn0HwyWtlPhkAz8gFokTNToPnb8Yla8M2Ha5MD1t64Sdaeu5935CgESszou0ROSl19G2NQs1Hsl7sFunfFHkdBoBn5CmgxRzcFWryfHPq9S5OS09eI2nP3846YAoVBuzL6LpGTUlmmGx0KtZvEfrkt0K077xa3KNZ0CYFmbNb/02IJ1Fo8juLx7yc1LtO05+7n3bD8CYN2ZexdoiCltmw7Ggp1NYj9cuuKxgp1W0KgGev1AaYt0J0Wi6CmDyeTMyTtuft5NyBQcsbeJQpSisvlx8Ki3vdivzwgUEOWEOicbI1zkQdboNkcPk2gG6YPJ5NXKdpz9/NOOPwJg3Zk5F2iJKW4vJIMEfvlfoGeMgwAgc4xhWkL1H6lkd1iKfP+dProRra4+cwfCdpz9/NOQKD0jLxLlCQUF4tAl099ZbaEQOf0IdCfflMbQz7+cv7P5+53b8/dz7vg9CcM2o1xd4kF8cU1uECPrJ5/z7LnFALN6UOg2+YcPJuhv9W9PXc/74DHnxBoJ0bdJSrEF1fPAgU2EQJtXgNNEGi2I3Qm+ErL9tz9vAM+gcKgXRh1l6gQXVzToECPQqA9ELGJtLhVk2AX/vtL1hLmfJHzl3/u3J67n7fH608YtAtj7hJVootrGhLoQWeCQOnpeBlT8nWg2/YKZrbK6ZnDJ7Tn7uftgUB7Ycxdokp8dUGgg9PtQvr8ULv45d9s/m1chbThn8OntOfu560J+BMG7cCIu0SN6OqaBgQ670wQKDlRt3IWgrQEmgk24V545+hx178Pn9Keu5+3BgLthxF3iRrR1TWFQAcn6jCRwpDOY1O7LoHmL7oXQVPac/fztgT9CYO2Z7xdok50dU39As07EwRKTcNxdvmVXyuZMw2BFoctJ1zA5dwAcloyvT13P29Jgz8h0NaMtksYxJeXV6Blb4JAielwoHJxJH3KVaDO2fd8nh4QaGR77n7ekiaBwqBtGW2XMIgvLwh0cJoe6bFT3HMwu12rKtC7Z4t3Ug6k33bt/2Q7Rc5t+JT23P28HY3+hEHbMtYuYZJQXx6BVjoTBEpLwkPllp/6P7PL5v+frTtXFo+KTzrLrk+BmreZmf1Q5vvxApX5+wW/bwhU3O+LfT+lwCDQgWl+rPG6+X/YOmknMfUh0J/nmL/M/C6R70f48+hRwb8f7w/2fhRN/nQatNVfAnOa/89UPjvOSeLzkJyXcAYEGtMeAsX7h+H9KBr9CYHSEvN/psA5quY5gE30IdAC87eJmYIF3o/yZz6Jl/j7Rb+PKbxLoA6Dxn65LYLl1RMvXHe03LFaLhDyEF8iov47t3fWnYrli6l/rk+Bmh1yBET6E9tIrRhll3CQUmDN/nQYNPbLfSOpE5+bLeMFevfK6uzF1Re/SAlTDpEThbvn7EQcuZj+OGPswteIFSgM2oZRdgkHKQXGItClpZOG/mIFemuxGb20cjUlUClEr7Ts336+EuzSiQuuc1QbgUCrRPsTBm3DGLuEi5QCi/CnbdDYLw89E6k+rowTaHGVecGplEiFkLZU/XDrswOub6UPPXMg0CoQaK+MsUu4SCkwLoEuLdWW86IEavoz+fG+Emi319ca3IlUIcGfMGgLRtglnKQUWIw/LYPGfrnjufB3PzlniXEa+UgPxwWS43usR4NAH/yPtAsTTiE23Qsf2Z67nycDgfbLCLuEk5QCi/KnadDYL7cFOmPvSiY/66D1BoHmo9SVqwfz2YcfZ+uDCQe7CaFBoLP/SJy09tja4zxIqek0psj23P08lSR/wqDpjK9LuEkpMAaBTqe3rAl4hECNs4Tzy80TryvnJ+o8ULrFXefs23HqfJv23P08kUR/QqDJjK5LeEgpsDh/GgaN/XKfQAuDLibxEQLdNOfs69ZCwBiIOZE+6bHvYXAifUmqQGHQVEbXJTykFBiPQHP5LYaPzQLNdpCeDL4yBmIeKke5sotnIuUk+xMGTWVsXcJHSn1F+rNu0Ngv9wvUfH5vs0Dnk9t6ix3i0dogJD3WmAA8lTMHAu2dsXUJHyn1FetPYoHmO0JP1v4ZFMemveKZ6WZkc/gIgZL+NwHPhc9o4U8YNJGRdQkvKfUVLdCjtALNVFEKsVGg2eTWmNuuj3AOP7RAszl5dVVzvqLpm8GntOfu50lAoP0zsi7hJaW8ov1ZNWjslwcEmsmvVGajQJ2jzZ0R7sNHPBee9tqsbAVz8kq+rPk486F7Dz6tPXc/T6GVP2HQNMbVJfyklBebQHdqe+iNAnUsgeYvjmwRNCxQY2WDhGxIOXn2zQMHPrpxaf4P91X0ie25+zkNtTggzy7o6RLxRKnTtGjsl4cEml3xWMzJGwXq3DByWlU4UY81Jl3XzdYw67xqvl3ZI2pov4C7n9NQjwMG7YCeLhEPm0CzOXkx1ooTqDlbn3+HKoFO97J9+ORzP0P89JrpQ3uPqLrJHm6/gLuf02DGAYO2Rk+XiIdNoNlYq1Bio0A3XVPb7DvGtQ3fdJhIfpayfWRqB8wx5WX7zdpVSsH2C7j7OQ1WHDBoW/R0iXgkCdTFS9WvOgwCXTwS6cgLn32+5ST9ZNB7lxY6PH2t9pZDoMH2C7j7OQ12HDBoS/R0iXjYBJptOBdvQqAZ+++cP3/e8zyPBW32zb75cG1mw7XXbxpvOAUaaL+Au5/T4Iij2uEZftFo0dMl4hmLQJ2XfOoTaH5iSg8C7QHufk6DKw4YtBV6ukQ8EOjAQKDCcMYBg7ZBT5eIh3kNFAKtA4EOjjsOGLQFerpEPJI2kbALD4EOjycOGDQdPV0inlSBHqW9DhQCrQOBDo4vDhg0GT1dIp5Uf84NGvvlhBfSHxKB7t/5LILrMm7/5+7nNPjjgEET0dMl4mETaJtbOQ/BnUhjgruf0xCIAwZNQ0+XiCfZnzODxn55SKDZwyuiDxNxClTfvfCjgruf0xCKAwZNQk+XiIdNoLWrmCJPY3IeJiJkQTAWCFQYwTiq/X6wXzRa9HSJeNL9eWDQ2C8nPFDZOVt3nFIvHghUGOE4YNAE9HSJeFr48+jR2C8PCNR4xiZOpB8h3P2choY4YNB49HSJeJgEaj67As9EGiHc/ZyGpjhg0Gj0dIl42vgz2qBNjzVeDB7xVE4n+1vzy5q2tvr6NZ3g7uc0NMYBg8aip0vEwyPQ7M7Nig7xXHiLhx+vVi+eP3Hhiz5/Vhu4+zkNzXHAoJHo6RLxtPJnrEF9Ar11zBiARgjUXDUtRrEjm8FHCnT/1nH7BqTlC7L2y7j7OQ0xccCgUejpEvF824bYL3cLdO9K5oPqgmaEQPO7HH+Rf2p/3fqSURAl0NueI0FpH/XRFe5+TkNUHDBoDHq6RDwDC3TrzrvF0KqmywiB5kPQpSMXv57Nb4/ZXzIKIgRanknv4KSg/2Bw93Ma4uKAQSPQ0yXi6V+gHhz7QQ02zJ9YWePp4Cck0ixQV5wLBA25ufs5DZFxwKDN6OkS8XAJ1JBljECnP1oT27HtIE1jnokU9Kckg3L3cxpi48BWUiN6ukQ8TAI9ZVggSqCWW06lRCqERoGuL+J76v3PZxcw7W/d/eScwP9qcPdzGqLjgEGb0NMl4uEQ6PIp67GScQIt9+/nrFxNCVQKTQLd8YW32FiScuEBdz+nIT4OGLQBPV0inoEFemT1/Huuh/LGCnQ6vXtlfnnk6gviLouMo0GgxYnKrv324r8eUibx3P2choQ4YNAwerpEPL0KFNg0CDT/b477vyUPjokagnL3cxpS4oBBg+jpEvFAoAPTcCL920FF7ogagnL3cxqS4oBBQ+jpEvFAoAMTFmg+xvRenbUZGJ8ODnc/pyExDhjUj54uEQ8EOjBhgWaC9B+Qki+Ryrj8lbuf05AaBwzqRU+XiAcCHZiwQM1TqtIbDAh3P6chOQ4Y1IeeLgHEEhRovgQa2CTaEbQIyt3PaUiPAwuhHvR0CSCWoECzGXrjs6GEHILK3c9paBEHDOpGT5cAYokQaEiPzS2Gg7uf09AmDhjUiZ4uAcQCgQqjVRwwqAs9XQKIRZFADzNVg3L/FgAODxGbSM1roNhEoqNlHBiD2ujpEkAsMZcxBS7z3MRlTMS0jgMGNdHTJYBYYi6k9w8w8+uccCE9He3jgEEN9HQJIJawQHcaTgtpen9QuPs5DR3igEHr6OkSQCxhgea3avo2iRreHhjufk5DlziwEFpDT5cAYok7zm7FqcjimSYylkDVVEuHD8OgVfR0CSCWyAOVVxznRe+dDR0WOjzc/ZyGbnHAoBX0dAkglqZHehSn+C+bj41aPOxYxhaSnmrp9HEYdIGeLgHE0iTQyoPzTrxfPv1k69PFU+VkXAQ61VMt3T4Pg5bo6RJALI1P5Swm8TmrB9ReELKDNNVTLR2/AAYt0NMlgFgaBVo++ciNlAXQqZ5q6fwVMGiGni4BxNIs0HKzyIVrc4kL7n5OA0UcMOgcPV0CiCVCoJX9IhNzZ4kV7n5OA0kcMOgMPV0CiCVGoAeD0CsufZ4UNPyc6qkWim+BQf8CgYIBiBPowSj0k3N1e564Kmn0OYO7n9NAFAe2kiBQMACxAp3x8M6n75w/4IX3r3/V3HpwuPs5DVRxwKAQKOifFIEKh7uf00AWBwyqp0sAsUCgwqCL49AbVE+XAGJpK9CHpL+CBO5+TgNhHIfdoHq6BBBLrEDv/O+1PaMHx5aOXBC2i8Tdz2mgjOOQG1RPlwBiiRLo7Cqm+i1HOwKvY+Lu5zTQxnGoDaqnSwCxRAh0P7sItHbsfHFI0y8EjUK5+zkNxHEcZoPq6RJALPH3wtdOrVsvLgcVcxaTnmqhWgU5MgAAIABJREFU/b5DbFA9XQKIpVGg5Vki1XPnK4fcuQ+r54C7n9NAHsfhXQjV0yWAWJoEujjNrjrWrJ5xJ2YMyt3PaaCP49AaVE+XAGJpEGhlqHmyJsqHHx8v3sCJ9JT0EMdhNaieLgHE0iDQ/LnFS8sX7fdu45lIPdBHHIfUoHq6BBBLWKDFANQ9TS/exVM5CekljsNpUD1dAoglLNB8B8m3zJkbVMhTPbj7OQ39xHEoDaqnSwCxhAW62TBHzwX7ku/9QeHu5zT0FcchNKieLgHEEhboetMUvbHBgHD3cxp6i+PwGVRPlwBiCQo0n6IHBpg7gq5k4u7nNPQXx6EzqJ4uAcQSFGh2uWdolz2bw8tYBOXu5zT0GMdhWwjV0yWAWCIEGtJjc4vh4O7nNPQZxyEzqJ4uAcTCJdBvPlybHLD2xs2UT31/aTJ51fcmdz+nodc4DpdB9XQJIJaOU/iWAr13aVJy+lr85zYmEGgnDpVB9XQJIJaOm0jZGmjiJtLj309qXI794PYEAu3IYTKoni4BxBJzGVPgZvfNFpcxmf6cTM7EfXB3AoF25hAZVE+XAGKJuZDeP0PPh6hpx4lsmP4MSLHK7iUIlIJDY1A9XQKIJepWTq8g87NGku5E2i2WPu9Pp49uZFJ85o/Nn7vXJFvufk7DAHEcFoPq6RJALFGHifi2kfJjQZP2kH76TU2Dj7+c//O5+w0fe/yPjaNV7n5OwxBxHBKD6ukSQCwNx9lthk6s2/tViwNBt00LZjP6t8Kf+mGxbgqBdudwGFRPlwBiaRBocfS86zzQW/nDPpIGoNkO0pngK/aHblQue4JACTgUW0l6ugQQS9MjPYoTlZeOXKxdq7RfHKecOACdXQpvLHnOF0V/+Wf/Z/7ltUydz/0dBErFYTConi4BxNL4ULny+ZsHw9CnLnw25+Pnjy9eTbuGadte8cxWRf1z+GLX/vL9DQiUjENgUD1dAoilUaCVpyI5SbuIPpuvGw7cCM/hN8obliBQQvQbVE+XAGJpfi582KCJNyE5R5u74X34mTaf+d39KQRKi3qD6ukSQCzNAp3uf+T356nEk0AdS6D5i/5F0ANtvnyz+F8QKB3aDaqnSwCxRAh0Ot0769bnyhepf865YeS06oKNl4sTmyBQYnQbVE+XAGKJEuiBQq8cs/R5Mlmfntn6fF4fczMSBEqNaoPq6RJALJECPeDhJ8+vFu48cuLC563+3LZrvyjbWWq4lH4GBEqOZoPq6RJALPECzdna6vIApD4E+vOcDj/rEFNdCOX+LQCMjWSBdgMCFQcMCkBrBhao85JPTOErDB+H2s14PV0CiAUCFQZDHFoNqqdLALFAoMLgiEOpQfV0CSCWtgJ9uLXV5mPYhW+CJQ6dBtXTJYBY2gi0OIlp+cJXqR+FQJtgikOjQfV0CSCWOIE+/OSFxaGfDyr3JYVu5fy+cohnYUgItAmuOBQaVE+XAGKJuRf+42PVM+l3avckrfgfGu8SKO5EaoItDn0G1dMlgFiaBZoPOIsHxz0w7un0PS8pQaAN98IvgED7RN1CqJ4uAcTSKNBiwJmfO28fbud9oodLoM6DlxpOY1oAgfaKNoPq6RJALE0CLQec+cHzO6Y//UfSuwTqnK07Tql3A4H2izKD6ukSQCwNAl0MOPOTk4snfKxc3SqeiuSfxNu0OZHeaAiB9ocug+rpEkAsDQJdPFPuwlygxYD0yfm/NsNDUBctnom0AALtG1UG1dMlgFgaBJoPOMuTk43nxG+GV0EdtHkqZwkE2juaDKqnSwCxhAWaDzjLBx8VM/pizJk/Nv4l3+dtWj0XvgACHQA9BmVPJRHx1QUGJyzQHWORs5jBl8bcrO7QRzGfw1ctuDGJncFDoIOgxqD8qaQhobrA0IQFum6sce6YVy7t1EeoEWQrnpNX8mXQx5k/Y/bgIdCB0GJQAakkIb64wOAEBZrP2Bcz9PUlQ5jZkDTp2cbZEHTy7JsHznx0I7vUKeYq+ikEOhRKDCohlRQkFBcYmgiBLmbw+ZJnZcqevZKyi5SvedZ51XzbvacEgQ6Ejq0kEakkIKG4wNAEBWrqsbio6SVvixh+es30p72nBIHyosKgMlLZnZTiAgOTJNBN6+bNNgK1xqCX7TchUGY0GFRIKjuTVFxgWFIEal7ENC3WQBMFOp3eq9zlefpa7S0IlPsXZCgwqJRUdiWxuMCQpAjUXgJts4mU8c2HazMbrr1+03gDAuX+BTnjN6iYVHYkubjAcKRsIu3Y975vJl/G1B/c/ZwGOXGM3qByUtkN7sICAWKuA32p9q+qLvNJfcrN8P3B3c9pkBTHyA0qKZVd4C4sECAs0M2qH70z+KQ7kfqDu5/TICqOcRtUVCo7wF1YIEDMvfD5nN2ewecD0JTz7HqEu5/TICuOURtUVirbw11YIEBYoLkhfzY7jKm4D34xgy925WUsgaqpFu5fUGPMC6HCUtka7sICARqOsyuOrzv52ZX8Ivpyvr5/65jxCjPc/ZwGaXGM2KDSUtkW7sICARoEWqx7LsgvavrxbfMVdrj7OQ3i4hivQcWlsiXchQUCND0TyXoGUr4lX3m4XMJpoL3C3c9pkBfHaA0qL5Xt4C4sEKDxqZy36v4sp+vr1ivccPdzGgTGMVaDCkxlK7gLCwRofi78rWMufxaro3L8qaZauH+BzUgNKjGVbeAuLBCgWaDT/Y8KfZaPRpoWc/sVGVcwzeHu5zTIjGOUBpWZynS4CwsEiBDoAXfeOX/+/Pu1q5VmVzWtXO3lN7WEu5/TIDSOMRpUaCqT4S4sECBOoA5+fOd9GZd/lnD3cxqkxjFCg0pNZSrchQUCtBaoPLj7OQ1i4xjfQqjYVCbCXVggAAQqDLlxjM6gclOZBndhgQAQqDAExzE2gwpOZRLchQUCQKDCkBzHyAwqOZUpcBcWCJAo0B/PrR5wQsjNm3W4+zkNouMYl0FFpzIB7sICAVIF2uYhcgPB3c9pkB3HqAwqO5XxcBcWCACBCkN6HCMyqPRUxsJdWCAABCoM8XGMx6DiUxkJd2GBABCoMOTHMRqDyk9lHNyFBQJAoMIYQRxjMegIUhkFd2GBABCoMMYQx0i2ksaQyhi4CwsEgECFMYo4xmHQUaQyAu7CAgEgUGGMI45RGHQcqWyGu7BAAAhUGCOJYwwGHUkqG+EuLBAAAhXGWOIYgUHHksomuAsLBIBAhTGaOOQbdDSpbIC7sEAACFQYI4pDukFHlMog3IUFAkCgwhhTHMINOqZUhuAuLBBAkUDB4FQMyv1TAOBAkUC5Bwo0jCsO0Quh40qlH+7CAgEgUGGMLA7JBh1ZKr1wFxYIAIEKY2xxCDbo2FLpg7uwQAAIVBiji0OuQUeXSg/chQUC4JEewhhfHGINOr5UuuEuLBAAD5UTxgjjkGrQEabSCXdhgQAQqDBGGYdMg44ylQ64CwsEgECFMc44RBp0nKm04S4sEKCbQPffOX/AC18T/ZhucPdzGkYah0SDjjSVFtyFBQKEBbp/57MD/H4UtSnP3c9pGGscAhdCx5pKE+7CAgHCAs0EufxB+H0IlJDRxiHPoKNNpQF3YYEAMQL1GxQCJWe8cYgz6HhTWYe7sECAKIEuLV0MvQ+BEjLiOKQZdMSprMFdWCBApECXng68D4ESMuY4hBl0zKmswl1YIECsQJdOubaSIFByRh2HLIOOOpUVuAsLBIgW6NITDoNCoOSMPA5JBh15Kku4CwsEiBfo0soXnvchUELGHocgg449lQXchQUCJAjUsRkPgZIz+jjkGHT0qczhLiwQIEagP/s/CoO+5H4fAqVj/HGIMej4U5nBXVggQJRA/7RZDEIvut/v8QfGw93PaVAQh5StJAWpnMNdWCBAnECnt445L2eCQMnREIcQg2pI5QzuwgIBIgU6fVAYtHY5EwRKjoo4ZBhURSr/AoGKJlag071iP6l6ORMESo6OOEQYVEcqIVDRRAt0+uNZ+3ImCJQcJXFIMKiSVEKgkokX6HT/betyJgiUHC1xCDCollRyFxYIkCDQ6f66eTkTBEqOljgEXM6kJZXchQUCpAh0Ov3IuJwJAiVHSxx/4TeollRyFxYIkCbQaXlB6NPu9znh7uc0aIljBrNBtaSSu7BAgESBLi4InV/OBIGSoyWOObwLoVpSyV1YIECqQBcXhM4uZ4JAydESRwarQbWkkruwQIBkgS4MuvInCJQeLXHkcBpUSyq5CwsESBfo4oLQn30BgZKjJY4CRoNqSSV3YYEALQRauSD0/4ZAqdESRwmfQbWkkruwQIA2Al1cEJqPRCFQOrTEsYDNoFpSyV1YIEArgVYuCG0r0G8+XJscsPbGzajmj//pb389a//s3/zH+7423P2cBi1xVGEyqJZUplcXGIyWAl1cENpKoPcuTUpOX2ts/sNrkwp/7XEudz+nQUscNXgMqiWVqdUFBqStQBcXhKYL9PHvJzUuNzT/cmLwqrMddz+nQUscdVgMqiWVadUFBqW1QBeXM6UK1PTnZHImrbnnA9z9nAYtcRhwLIRqSWVSdYFhaS/QhUETBbphC9E9pPQ2d3+Au5/ToCUOEwaDakllUnWBYekg0One2TYC3S2WPu9Pp49uZIuhz/yxqfkzL//zwT8ef/NhbtC37Jbc/ZwGLXFYDG9QLalMqS4wMGGBTh9uHfCV7938gtAkgf70m9oQMl/gfM63t543X7yfbyg5PsDdz2nQEofN4AbVksqE6gJD0yDQBvY/Pn/AC183tyzZNqfgG74R5ZxsAHqmost8TdT+AHc/p0FLHA6GNqiWVCZUFxiabgJNJ9PfmeArVeZ6/eWfqy99f8m9Csrdz2nQEoeLgQ2qJZWE5QeoGVqgc/vVlzx3bUeWZDN4Q5Yb7jk8dz+nQUscbgY1qJZUEpYfoGZogW7b8ssk6Z7DO3ybGxcCHSVDGlRLKgnLD1AzsECz+bprROmewztHp54hK3c/p0FLHD4GNKiWVJJVH6BnYIE6R5ueEeWcR//6h9+ab2EEOmaGM6iWVJJVH6BnYIE6p+TzFz2LoC623SNW7n5Og5Y4/Ay2laQllV2LDvTIwAJ1zr6dVvXjXAaY6qkW7l/QO0MZVEsqO1cd6A8OgZqz7/m8Pl6gu557l7j7OQ1a4ggxkEG1pLJz1YH+GFigztl3NqT0XUpvki2j4k6kMTOMQbWksnPVgf4Yn0CtO5d+nkP1G0H/VA3K/VsAaM/oBJrdClr9Cgh0hMCgQAUDC9R5yWeKQDN/OrfsuWdaNGiJo5H+Z/FaUtm56kB/jEyg24HT77j7OQ1a4mimd4NqSWXnqgP9MSqBFg/3cDfl7uc0aIkjgr4NqiWV3csO9MaY1kCLh3t4zq/n7uc0aIkjhp4347WksnvZgd4YkUCLZ3O+6Xmfu5/ToCWOKPo1qJZUdi870Bs9CvT7yrOLC0N2EGj+dc9423H3cxq0xBFHrwbVksrOlQj6Y2CBtr8TKds+CjXj7uc0aIkjkj4NqiWV3UsR9IYEgUbcC19sH52+6W/D3c9p0BJHLD0aVEsqCWoR9MXAAnUevNR8GlOx/Ol99twM7n5Og5Y4ounPoFpSSVCLoC8GFqhztu44pd79TZeDf4+7n9OgJY4E+jKollR2rkTQH8JPpJ9TbB/5tt9zuPs5DVriSKEng2pJZfeyA70h/JlIMwp/Np13x93PadASRxL9GFRLKinqDvSE8KdyTkt/nm48sZ67n9OgJY40elkI1ZJKiroDPSH9ufDF7UfB7aMM7n5Og5Y4EunDoFpSSVF3oCeGFmh+PWdlFdQ637NO9vaZZn+qqRbuX8BDDwbVkkqKsgM9MbhAsxXPySu5ER9vhAeYu7HjTz3Vwv0LmKA3qJZUUpUe6IHBBVrcUvTsmwdSfHSjYYMom8DHPbGTu5/ToCWOdMgNqiWVhMUHqBleoMWZSlVeNd8ulLlrt82xpcrdz2nQEkcLqA2qJZVDVSZowfACnf70milDe0+p0OMGBHqYoDWollQOVJegDQwCtcagl+03cz3mC6YQ6GGB1KBaUjlMVYJWcAh0Or1Xucvz9LXaWzWBmneDQqDaoTSollQOU5OgFTwCnU6/+XBtpsG1183jlWoC9S+BQqBKITSollQOUY+gJVwC7QHufk6DljhaQ7eVpCWV3IUFAkCgwtASR3vIDKolldyFBQJAoMLQEkcHqAyqJZXchQUCQKDC0BJHF4gMqiWV3IUFAkCgwtASRydoDKolldyFBQJAoMLQEkc3SAyqJZXchQUCQKDC0BJHVwgMqiWV3IUFAkCgwtASR2e6G1RLKrkLCwSAQIWhJY7udDaollRyFxYIAIEKQ0scBHRdCNWSSu7CAgEgUGFoiYOCjgbVkkruwgIBIFBhaImDhG4G1ZJK7sICASBQYWiJg4ZOBtWSSu7CAgEgUGFoiYOILgbVkkruwgIBIFBhaImDig4G1ZJK7sICASBQYWiJg47WBtWSSu7CAgEgUGFoiYOQtgbVkkruwgIBIFBhaImDkpYG1ZJK7sICASBQYWiJg5R2C6FaUsldWCAABCoMLXHQ0sqgWlLJXVggAAQqDC1xENPGoFpSyV1YIAAEKgwtcVDTwqBaUsldWCAABCoMLXGQk25QLankLiwQAAIVhpY46Ek2qJZUchcWCACBCkNLHH2QaFAtqeQuLBAAAhWGljh6Ic2gWlLJXVggAAQqDC1x9EOSQbWkkruwQAAIVBha4uiJFINqSSV3YYEAEKgwtMTRFwlbSVpSyV1YIAAEKgwtcfRGvEG1pJK7sEAACFQYWuLoj2iDakkld2GBAIoECg4JVYOGGw70g8DhRZFAuQcKNGiJo08ix6ApB49IhruwQAAIVBha4uiVOINCoKB3IFBhaImjZyIMmnDXvGy4CwsEgECFoSWOvmk2KAQK+gcCFYaWOHqnyaAJN80Lh7uwQAAIVBha4uif8EJo/B2f4uEuLBAAAhWGljgGIGjQ2NuVRgB3YYEAEKgwtMQxBAGDxl5sPwa4CwsEgECFoSWOQfAbFAIFgwCBCkNLHMPgM+hRv1rHB3dhgQAQqDC0xDEQHlFCoGAYIFBhaIljMFymPHpUk0G5CwsEgECFoSWO4bBNeRQCBQMBgQpDSxwDYqnSFOjIDcpdWCAABCoMLXEMieFKy58jNyh3YYEAEKgwtMQxKHVXQqBgMCBQYWiJY1gcztRjUO7CAgEgUGFoiWNgIFDAAgQqDC1xDI1ig3IXFggAgQpDSxyDA4ECBiBQYWiJgwGtBuUuLBAAAhWGljg4UGpQ7sICASBQYWiJgwUIFAwMBCoMLXHwoNKg3IUFAkCgwtASBw8QKBgWCFQYWuJgIbwIOlaDchcWCACBCkNLHBw0+BMCBeRAoMLQEgcHTQIdqUG5CwsEgECFoSUOBhr9OVKDchcWCACBCkNLHAxAoGBwIFBhaIljeCL8OU6DchcWCACBCkNLHMMDgYLhgUCFoSWOwYny5ygNyl1YIAAEKgwtcQxNpD8hUEAKl0C/+XBtcsDaGzejmv/w9/Pmz7587b63DXc/p0FLHEMTK9ARGpSq5kAP8Aj03qVJyelrjc1/eG3RfHLZp1Dufk6DljgGJtqfIzQobe0BUjgE+vj3kxqXG9pv15s/86a7GXc/p0FLHAMDgQIWGARq+nMyORNsv2E2n7zqbMfdz2nQEsewJPhzfAbtowgBEQwCtYXoMWLGtt188parIXc/p0FLHMMCgQIehhfobrH0eX86fXQjWwx95o/e5t/nLX733w/+8W9fZv/65Z8dLbn7OQ1a4hiUJH+OzqC9lSLozuAC/ek3tTHn4y/n/3zOtzOUz/fL9/N/u4as3P2cBi1xDEmiPyFQQMbgAt02Bbjhn5RPi/Fqxa+ZQV3C5e7nNGiJY0hSBToyg9IWICBlaIFm/jsTfKXKhjXD3/XN+bn7OQ1a4hiQZH+OzKCE5QeoGVqg8yXNuv52vauaTrvO1wAgUFACgQI2hhbotj0Bz1ZF3XP4bAvpLas5BAoKWvhzXAYlLD9AzcACzUaUxhbQRmAO//hf//63/742OoVAQQ0IFPAxsECdo83d4D68ibc1dz+nQUscg9HKn6MyaOeqA/0xsEAdS6D5i55FUItMwbiMCTSiJZWdqw70x8ACdW4YOa3q4V8ueYer3P2cBi1xCEBLKruXHegNDoGa+vOuapr8241f+29EUlMt3L9ADVpSSVF4oCcGFui2a78o21nyXUqfk9/BFFgu5e7nNGiJQwBaUklTeqAXRiPQ74sjRK3T7H6eQ/k7AQCgmdEIdLccgD5jHKkMgQIAeBhYoM5LPhMF6jtSmXumRYOWOASgJZVEtQf6YDQC3f7r1//Tf/nDa7lCcRkTaERLKqmKD/TAaARatP3Sa1Dufk6DljgEoCWV3csO9MZo1kBL7k08xzFx93MatMQhAC2p7Fx1oD/GJ9D8RFH73nnufk6DljgEoCWVXYsO9EiPAv2+8uziwpAUAs2a2xfTc/dzGrTEIQAtqexaiKBHBhZotzuRcjxHKnP3cxq0xCEALansWoigRyQINOVe+MUXWyNW7n5Og5Y4BKAllV0LEfTIwAJ1HryUdBrT1HsgE3c/p0FLHALQksquhQh6ZGCBOmfrjlPqSx795z/82lwzhUBBDFpSSVSPoA+En0i/7dovwhQexKAllWTVB+gR/kyk3YnjPWwigRi0pJKw/AA1wp/KmQ0266NTXMYEotCSSsr6A8QIfy589qZDuN4HyY8dnCpFBlIJemdogeb3EVVWQTdcs3SjeXW4uXvJPYNXAqqeDKQS9M7gAs1Pln8lXwZ9nPnT+0zOvPnpm8ULmVHVDkBR9XQglaB3BhdoYcBn3zxw5qMb2aVOgfFkcQ7oy/88nbXPHoqUctXoyEDVk4FUgt4ZXqDZsmadV823K4LcsJvrncCj6glBKkHvDC/Q6U+vmT6095SqI8wvD5M/UfV0IJWgdxgEao1BL9tv1qbo24Y/FwuiCkHVk4FUgt7hEOh0eq9yl+fpa7W3XFd5/vDbij6fdT4QSQ2oejKQStA7PAKdTr/5cG1mw7XXzdGk+zL5x//02/kV+GtvaB59AgDGBZdAAQBg9ECgAADQEggUAABaAoECAEBLIFAAAGgJBAoAAC2BQAEAoCUQaG+UN1y5Tz4pTknJDqLanTRgPQLqMIFUAqFAoL2xuGPVedhpcUhKm6ovbkM4LLcV9JjKjNmTD+BVkA4E2huLqneVZn7QaZuqD9wIq5TeUlmw4flqAMJAoL2xqHrXcdG79Tfjqz54FItSekplybbHqwA0AIH2xqLqXcfvlUdMJVa9fZyq2sP5F/STypJdj1cBaAIC7Y2s6tfctTl/c21R9XU2fG84z5fWX/n9pLIge8rWIUgjoAcC7Y2s6v/mkrOAZ7sWz/xtetXnw6vT1+IeiKKEXlJZcO/Q/HcI0AOB9kZW9a/83qm42bTzuX9Mrvp8v6Scz385afSDBvpIZfHV/3h4BvKAHgi0N7Kqf9W5QTF/L3srqeoTHwqthT5SmfHDYkkZAgXpQKC9kVf9966J53za+cfkqs++0n6ElPZ9pB5SmX3vjcoVYRAoSAcC7Y286uf/n3kHzXzaeT+56ucGqU9i54uiep/ynNFDKmf8y2uZOp/7OwgUtAMC7Y286rNp91v2W/k7KVXv+EC2Kqp8Dt9DKqeLCxou39+AQEE7INDeKKp+PmysT7Jn1jsYSKZWffGNdlvlc3j6VJZvze/lgkBBSyDQ3iiq3jHx3M2KOrXqnaPN3fAuiQroU1m89czv7k8hUNAaCLQ3yvHihqW9jWwglVr1jiXQ/EXli6D0qczfevnmohEECtKBQHujrPpdc5I9H0m+lV71zg0jp1WVQZ/K+Vsv36w2gkBBOhBob5RVPy/yqvd283+3Eqj58vzbD4tAyVLpaASBgnQg0N5YbPlsGIrLp53JVb/t2i/K/ozubXj6VDoaQaAgHQi0NxZVv1uvz2LaCYHGQp9KRyMIFKQDgfbGournZb4o4mLaCYHGQp9KRyMIFKQDgfZG5arN+sRzu9BgatU7L/k8XAKlSqWjEQQK0oFAe6NS9bWJ5/z1ufAg0EjoU+loBIGCdCDQ3qhUfe0UjOz0i9n/gkAjoU+loxEECtKBQHujUvWPqydZbpcVjTXQSOhT6WgEgYJ0INDeqN65XjkFIz/9onwVAm2GPpWORhAoSAcC7Y1q1VdOwVhMOyHQWOhT6WgEgYJ0INDeqFZ95RSMxbQTdyLFQp9KRyMIFKQDgfZG7fC5cuJZmXbSCPRQ3Qs/gySVjkYQKEgHAu2NWtWXE8/8/Mo5rU5jch4mclhOY5pBkkpHIwgUpAOB9kat6suJ526lnludB2oONr1fogj6VDoaQaAgHQi0N+rnxxcnWW5UXsOJ9JHQp9LRCAIF6UCgvVGv+vwOmvL0ixl4JlIkPaTSbgSBgnQg0N6oV31+CkZ5+sUMPJUzkh5SaTeCQEE6EGhvGBPu7BSM4vzKOXgufCQ9pNJuBIGCdCDQ3jCqfj5U/J+r084WVT//RLXUsydLKp/B95JKqxEECtKBQHvDqPpssbI2306v+vxLXsnfepz5U/kefD+ptBpBoCAdCLQ3zD3zjbzqF/PtFlWfDUEnz7558OajG5fm/1B+Ff20p1SajSBQkA4E2htm1e/mVb+Yb7eo+uxL6+iv/F5SaTbSn0ZADwTaG2bVf3/JmHa2qvqfXjP9qX0HadpXKo1GEChIBwLtDbPq88FjpZhbVb05Br1M/bsF0lMq640gUJAOBNob1n1D1hZ6y6q/d2mhz9PXSH+zUHpLZbURBArSgUB7w6p66zL41lX/zYdrsxZrr98k/L2C6TGVi0YQKEgHAgUAgJZAoAAA0BIIFAAAWgKBAgBASyBQAABoCQQKAAAtgUABAKAlECgAALQEAgUAgJZAoAAA0BIIFAAAWgKBAgCUIeo1AAAKL0lEQVRASyBQAABoCQQKAAAtgUABAKAlECgAALQEAgUAgJZAoAAA0BIIFAAAWgKBAgBASyBQ5ewszfjZn0JtHhybN3ppqN8U8/eG/k0AtAECVY5boPsf/V+Vf7llVW9DCQQKtACBKscp0FvHamZyyspoQwkECrQAgSrHIdC9s4aZHLKy2lACgQItQKDKsQS6/5FlJktWjjaUQKBACxCociyBOsxkvdSzvSBQoAUIVDkxArUQIFAAxgAEqhwIFID+gECVA4EC0B8QqHIgUAD6AwJVDgQKQH9AoMqBQAHoDwhUORAoAP0BgYrh9vPHD6yyfOK9rysv7t8+N7PNkRNXv/Z+MAi5QPfvvHPu+PztI6sveH/Vw0+yRsuuNp0FuvfuapaVC181tVkOtQGgIxAoJzulSPY/zqQy51ShnPyOIOPVykftU5bW568/8bWr3f7bSwZPz9tUheZrk/+g22eNd1c+t8Pav3W83uZq/X2nQHfyxk9+7Wyzs/gttW83v7v4BceqbT4vErP8gas1AG2BQDkpBbpX99JK5sUHx+qvfmF9dGCB3jpmvnnAyS+mdR6YkjV+ulOgO/lX5/70C3TP/PaT9ijY+gUH//GBQEEPQKCcFAJ9YIppLrwd56u1jw4qUPutDMNKt5xtGg4vMf3pFaiVqvI/Nws27b+/8icIFPQABMpJLlCHFJ60xp8zFl5kEKjPn8av2PE0qujSFugD058+gTqSUkvLDIc/Dwz6P83+XwgU0AKBcpJZ4aLDTMsf/PiroIWGF+i6y0sZTy7+vvNnGz/Vd3hJxZ8egf7CLfH6YoD/d0KggBgIlJOs1lfn/+/KbLd6sUnz7+aqWD452/94WO4wPWl8dECBll5aeW9r/u1bn55zyLHQ7In3t2b/fHj7ivE9U1uODn96BJpr8OTn81yVf786BF0YPG91pfpJCBSQAoFysrDCYjPZufFeaG1hqjYCndPyOLviF9QUtFfYtmyZ66vaqtj1ecI3PXf5MyTQX5hfVPulhcGfKPet9hb/TYBAAS0QKCeLUV3lwONFuVf2v4txlSGU4QS6Yxq8+ucWI+Mdh6csqda/3ulPv0Br31181hrd1i6+WvxHCQIFtECgnOy46rpUxZPVppuGF4YW6Lo11JuTy7H8e5v2Ly+/7mnj39l37f3KEW1AoM6rRxfxrjv8uRiWQqCAFgiUk8IKtXIvBpv1Ys+FYgz1BhNo/qMMy00LY5bfv+7SVz6oLj9c/XqPP70CNdrlv6v8+/nHjH159wIEAJ2BQDnZcZV1Uex1VZgGG1ig3rsvje9fd9pw88iJF69vub7sR48/fQK1DLhe//v5QN36nb6PA9AJCJQTawI6xzlbNkdxg28i3f30/Kr95/KmhkAbPLX4+sKfpwJtqnFYQ0v3CNhqVaQPAgW0QKCc5FZwLtgZtc4tUB+GQPMR4PLF5s8cfP2PZ33+9AnUGqnWBZoL+WmzVdEOAgW0QKCc7DgnnMagKseYHAsVaHlh0fLJ6w2fWXqpWKxw+NMnUEuN9W3/7FMuTfrfAaA9ECgnrot+xiTQ/TtXjtW/v3Yd/pEXHUc1Lb7+onUVaeA3uf9bY6bQl5VibAqBAlogUE5CAjXX8YQJdOvOx88v7ktffL95s7rrIFOzjUt4boHaAqy/7k7djMzsECigBQLlZIwCffjZu8+vLhlUvt8+yuPIBSMU60QQ++qolgL17SEVb0GggBYIlBO3BeUKtHL7uVegzsOQ6sceWwJ1eA0CBWMAAuVkXAKtnZrvF6j72OXqkcoVgeZjWVt5ECgYAxAoJ6MSqOOgebdArUd6ZI6zT69bWnq62HSydtchUDAGIFBOxiRQ63z8GSeu/r/HjO/PeGiPVq3DRObazP+3ZTYIFIwBCJSTEQnUXLc88tSL178uX3fto+9/YgxZyx9VOzFp3XjT8xNSduGdvwa78KAHIFBOxiPQyvWdqy98tmWeHO9S1oy771Yn88YBoNm0vbib05jEUwsU14GCPoBAORmPQIvDkKynCPsvXS+wD46vf737nI92AvW1wp1IoB8gUE46C9QSQj5SpBZoMQC1VxebBbo4Er5oZXy9cxLfTqD+e+H9agWgPRAoJ/QCNQ84JhKoeeym+WPDAjXPQjK+3jmJbydQ678fJVHHRAGQCATKSWeBWte6W6e+0QjUd5RHYazy+2d3eB73jVM9AnVO4tsJ1Hvokm+zH4BOQKCc0AvUmg6TCtR3nrJxGZE9Hg0L1LVA0FKgnhPpi7xAoIAWCJSTDgI1H/GRUcyGhxqBmo8fMZ/c5P4Vvr9X/VxLgXqeiVTcXgqBAlogUE46CNTx/OCyUW8CNQ/9KC9uMsaWZkTG0qT9EzbrX+NoEytQh4un01tLECjoBQiUkw4CdU179wt/hgTq2Kj27RnZDwuuz+H3yivljU0c88cbA1P/ZabeZ8dHC7T8b8ip8qv2y6caQ6CAGAiUkw4CLWeli3YLn0UItDqY9Al00aZ8qmXlUR21s0WKz5bXi1bODtn/B8NegRud3I8+niYItFhVWFo6+fnXs2c5Xan8TAgU0AKBctJFoJWHZ8zOfX94O9PnytkGgZbP0TDvJnJc9rloU55Rt/Le/OmaW5/WD7Z7yfhk2fDh3XeLH2qu39aGs/kfMJ8Nki7Q0uEuIFBACwTKSReBlnPVuiH+w9uRAq18nVegizaLcZ3ByaypOXJ0YJ1aXxOo+Xzj9gJ1Hki69CLuRAI9AIFy0kmgTqflD2oLCLTml6ydd1On0sYzrnvau7RgEtggmlb/wEvuNgkCdf2Ei7iVE/QBBMpJJ4E6BnvLF6fNAq1+LHvdFprdxinG5auOW5HcBq2qy31e83rtz3UR6PT2MeuPQ6CgDyBQTroJtLprNGflg2mEQKuGy4TiEJrVxnXQ/KmvnR92nbx8qhqNW6D1SXwngdY3uJZnfxz3woM+gEA56SjQA6lVTotbnj+7LUKgVRXOBeUSmtnmgL0rNSeezDba8/XS6g+zTqQ/+UX1q32PXKpN4rsJ9IC776zOvuLIU1crSxAN9+wDkAgEOnL23i08kfCh/dvnZh9aXn3hi6Q2+7evzB9idPDi9Ya/8On51WNxTYcBAgV9AIGCQ4H/ifEAtAcCBYcC5woIAB2BQIEmNmfPavrKft1/0jIAHYBAgSZ2PDN173l8AHQBAgWaCB/Ujz0kQAwECjThPia1uDUVS6CAGAgUaKI8F696fVZxYxIuowfUQKBAFeU9+yfen58GtX/34/KqfgxAATUQKFBF7SCpOrgIFJADgQJdeM/Tww4SoAcCBcrwGBTjT9ADECjQRuUZSCXLf8X9q4BKIFCgD+PcqPygKgDIgUCBSu6+ey7bfV996j3HrZ0AkACBAgBASyBQAABoCQQKAAAtgUABAKAlECgAALQEAgUAgJZAoAAA0BIIFAAAWgKBAgBASyBQAABoCQQKAAAtgUABAKAlECgAALQEAgUAgJb8/zhY5/g28vhSAAAAAElFTkSuQmCC)
Simulo un singolo (piccolo) dataset
N = 16
n = N/4
id = 1:N
MT = c( rep(0,n), rep(0,n), rep(1,n), rep(1,n) )
DY = c( rep(0,n), rep(1,n), rep(0,n), rep(1,n) )
id; MT; DY
## [1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
## [1] 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1
## [1] 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1
# determino lo score "atteso" per ogni soggetto
score_atteso = intercept +
MT * multitaskingEffect +
DY * dysfluencyEffect +
MT * DY * interactionEffect
cbind(id, MT, DY, score_atteso)
## id MT DY score_atteso
## [1,] 1 0 0 0.00
## [2,] 2 0 0 0.00
## [3,] 3 0 0 0.00
## [4,] 4 0 0 0.00
## [5,] 5 0 1 0.39
## [6,] 6 0 1 0.39
## [7,] 7 0 1 0.39
## [8,] 8 0 1 0.39
## [9,] 9 1 0 -0.28
## [10,] 10 1 0 -0.28
## [11,] 11 1 0 -0.28
## [12,] 12 1 0 -0.28
## [13,] 13 1 1 0.31
## [14,] 14 1 1 0.31
## [15,] 15 1 1 0.31
## [16,] 16 1 1 0.31
# determino e aggiungo la varianza residua
residual = rnorm(N,0,sigma)
score = score_atteso + residual
# metto tutto nel dataset
df = data.frame(id=id, multitasking=MT, dysfluency=DY, score=score)
df
## id multitasking dysfluency score
## 1 1 0 0 -0.14180297
## 2 2 0 0 0.85175640
## 3 3 0 0 -0.33599124
## 4 4 0 0 -0.41089111
## 5 5 0 1 0.68914635
## 6 6 0 1 -1.21302280
## 7 7 0 1 1.07210925
## 8 8 0 1 0.68658147
## 9 9 1 0 -0.25401097
## 10 10 1 0 -0.67662425
## 11 11 1 0 0.69632704
## 12 12 1 0 0.02401117
## 13 13 1 1 0.06071607
## 14 14 1 1 1.20459748
## 15 15 1 1 -0.18771538
## 16 16 1 1 0.54085832
anche se inutile qui, fitto e guardo il modello
fit1 = lm(score ~ multitasking * dysfluency, data=df)
summary(fit1)
##
## Call:
## lm(formula = score ~ multitasking * dysfluency, data = df)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.52173 -0.35834 -0.02799 0.47256 0.86099
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.009232 0.363356 -0.025 0.980
## multitasking -0.043342 0.513863 -0.084 0.934
## dysfluency 0.317936 0.513863 0.619 0.548
## multitasking:dysfluency 0.139253 0.726713 0.192 0.851
##
## Residual standard error: 0.7267 on 12 degrees of freedom
## Multiple R-squared: 0.0895, Adjusted R-squared: -0.1381
## F-statistic: 0.3932 on 3 and 12 DF, p-value: 0.7602
Simulo 2000 volte
ora faccio la power analysis (con N tot = 320) e vedo anche le stime
ottenute
# decido il numero di iterazioni
niter = 2000
# stabilisco un N un po' decente
N = 320
n = round(N/4)
# inizializzo (per praticità) il vettore dei pvalue, dei CI e degli effetti stimati
pvalue = rep(NA,niter)
ci_width = rep(NA,niter)
estimated_interactionEffect = rep(NA,niter)
# lancio il ciclo for
for(i in 1:niter){
# simulo i punteggi
id = 1:N
MT = c(rep(0,n),rep(0,n),rep(1,n),rep(1,n))
DY = c(rep(0,n),rep(1,n),rep(0,n),rep(1,n))
score_atteso = intercept +
MT * multitaskingEffect +
DY * dysfluencyEffect +
MT * DY * interactionEffect
residual = rnorm(N,0,sigma)
score = score_atteso + residual
# metto tutto nel dataset
df = data.frame(id=id, multitasking=MT, dysfluency=DY, score=score)
# calcolo il p-value dall'anova
fit1 = lm(score ~ multitasking * dysfluency, data=df)
fit0 = lm(score ~ multitasking + dysfluency, data=df)
pvalue[i] = anova(fit1,fit0)$"Pr(>F)"[2]
# estraggo il coefficiente di interazione stimato e lo salvo nel vettore
estimated_interactionEffect[i] = fit1$coefficients["multitasking:dysfluency"]
# estraggo i CI e li salvo nel vettore
ci = confint(fit1)
lower = ci["multitasking:dysfluency",1]
upper = ci["multitasking:dysfluency",2]
ci_width[i] = (upper-lower)
}
Calcolo il power
mean(pvalue < .05, na.rm=T)
## [1] 0.1575
Determino il minimo effetto significativo
minsign = min( estimated_interactionEffect [pvalue < .05 & estimated_interactionEffect > 0], na.rm=T )
minsign
## [1] 0.3878323
Guardo la distribuzione degli effetti stimati e dove si colloca
quello minimo significativo (linea tratteggiata rossa) e quello “vero” (linea blu)
hist(estimated_interactionEffect, breaks=30)
abline(v=minsign, col="red",lwd=4,lty=2)
abline(v=interactionEffect, col="blue",lwd=4,lty=1)
![](data:image/png;base64,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)
Do anche un’occhiata alla distribuzione dei CI
hist(ci_width/2, breaks=20)
abline(v=mean(ci_width/2),lwd=8,lty=1)
![](data:image/png;base64,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)